To see the other types of publications on this topic, follow the link: Learning long short-term memory.
Dissertations / Theses on the topic 'Learning long short-term memory'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Learning long short-term memory.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Cumming, N. "The Hebb effect : investigating long-term learning from short-term memory." Thesis, University of Cambridge, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.598214.
Full textAbstract:
How do we learn a sequence of items so we can remember it not only over the short-term, as in hearing a phone-number and repeating it back, but over the long term? Ten experiments are presented that investigate this problem using the Hebb repetition effect (Hebb, 1961). In a canonical Hebb effect experiment, lists of familiar items are presented in an immediate serial recall task and one list is repeatedly presented at regular intervals. This leads to an improvement in recall for the repeating list over baseline performance. Existing models of serial order learning are tested; Chapter 2 provides evidence contrary to positional models of the Hebb effect while Chapter 5 provides evidence against chaining models. The experiments in these chapters (Experiments 1, 2 and 7) use a transfer design where a representation of the repeating list (Hebb list) is built up, then a list is presented whose structure is derived from the Hebb list in a way that tests the predictions of these models. The experiments of Chapters 4 and 5 examine the hypothesis that the most parsimonious model of the Hebb effect is one that is based on the formation of chunks (Miller, 1956), higher-level representations of several items. The results of these experiments are consistent with a chunking model based on the Primacy model (Page and Norris, 1998), but do not provide direct evidence of a chunking process. A growing body of evidence (e.g. Baddley et al., 1988; Papagno et al., 1991) suggests that the phonological store of the working memory model (Baddeley, 1986) plays an important role in the development of long term representations required for the acquisition of new vocabulary. For example, the ability to learn new words is impaired in patients with damage to the phonological store (e.g. PV, SC) and in normal subjects performing articulatory suppression. In chapter 6, the hypothesis that the Hebb effect is an experimental analogue of phonological form learning is investigated, the results of which suggest that the Hebb effect is involved in at least some of the same processes.
2
Bailey, Tony J. "Neuromorphic Architecture with Heterogeneously Integrated Short-Term and Long-Term Learning Paradigms." University of Cincinnati / OhioLINK, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1554217105047975.
Full text
3
Shojaee, Ali B. S. "Bacteria Growth Modeling using Long-Short-Term-Memory Networks." University of Cincinnati / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1617105038908441.
Full text
4
Gustafsson, Anton, and Julian Sjödal. "Energy Predictions of Multiple Buildings using Bi-directional Long short-term Memory." Thesis, Högskolan i Halmstad, Akademin för informationsteknologi, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:hh:diva-43552.
Full textAbstract:
The process of energy consumption and monitoring of a buildingis time-consuming. Therefore, an feasible approach for using trans-fer learning is presented to decrease the necessary time to extract re-quired large dataset. The technique applies a bidirectional long shortterm memory recurrent neural network using sequence to sequenceprediction. The idea involves a training phase that extracts informa-tion and patterns of a building that is presented with a reasonablysized dataset. The validation phase uses a dataset that is not sufficientin size. This dataset was acquired through a related paper, the resultscan therefore be validated accordingly. The conducted experimentsinclude four cases that involve different strategies in training and val-idation phases and percentages of fine-tuning. Our proposed modelgenerated better scores in terms of prediction performance comparedto the related paper.
5
Yangyang, Wen. "Sensor numerical prediction based on long-term and short-term memory neural network." Thesis, Mittuniversitetet, Institutionen för informationssystem och –teknologi, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-39165.
Full textAbstract:
Many sensor nodes are scattered in the sensor network,which are used in all aspects of life due to their small size, low power consumption, and multiple functions. With the advent of the Internet of Things, more small sensor devices will appear in our lives. The research of deep learning neural networks is generally based on large and medium-sized devices such as servers and computers, and it is rarely heard about the research of neural networks based on small Internet of Things devices. In this study, the Internet of Things devices are divided into three types: large, medium, and small in terms of device size, running speed, and computing power. More vividly, I classify the laptop as a medium- sized device, the device with more computing power than the laptop, like server, as a large-size IoT(Internet of Things) device, and the IoT mobile device that is smaller than it as a small IoT device. The purpose of this paper is to explore the feasibility, usefulness, and effectiveness of long-short-term memory neural network model value prediction research based on small IoT devices. In the control experiment of small and medium-sized Internet of Things devices, the following results are obtained: the error curves of the training set and verification set of small and medium-sized devices have the same downward trend, and similar accuracy and errors. But in terms of time consumption, small equipment is about 12 times that of medium-sized equipment. Therefore, it can be concluded that the LSTM(long-and-short-term memory neural networks) model value prediction research based on small IoT devices is feasible, and the results are useful and effective. One of the main problems encountered when the LSTM model is extended to small devices is time-consuming.
6
Singh, Akash. "Anomaly Detection for Temporal Data using Long Short-Term Memory (LSTM)." Thesis, KTH, Skolan för informations- och kommunikationsteknik (ICT), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-215723.
Full textAbstract:
We explore the use of Long short-term memory (LSTM) for anomaly detection in temporal data. Due to the challenges in obtaining labeled anomaly datasets, an unsupervised approach is employed. We train recurrent neural networks (RNNs) with LSTM units to learn the normal time series patterns and predict future values. The resulting prediction errors are modeled to give anomaly scores. We investigate different ways of maintaining LSTM state, and the effect of using a fixed number of time steps on LSTM prediction and detection performance. LSTMs are also compared to feed-forward neural networks with fixed size time windows over inputs. Our experiments, with three real-world datasets, show that while LSTM RNNs are suitable for general purpose time series modeling and anomaly detection, maintaining LSTM state is crucial for getting desired results. Moreover, LSTMs may not be required at all for simple time series.<br>Vi undersöker Long short-term memory (LSTM) för avvikelsedetektion i tidsseriedata. På grund av svårigheterna i att hitta data med etiketter så har ett oövervakat an-greppssätt använts. Vi tränar rekursiva neuronnät (RNN) med LSTM-noder för att lära modellen det normala tidsseriemönstret och prediktera framtida värden. Vi undersö-ker olika sätt av att behålla LSTM-tillståndet och effekter av att använda ett konstant antal tidssteg på LSTM-prediktionen och avvikelsedetektionsprestandan. LSTM är också jämförda med vanliga neuronnät med fasta tidsfönster över indata. Våra experiment med tre verkliga datasetvisar att även om LSTM RNN är tillämpbara för generell tidsseriemodellering och avvikelsedetektion så är det avgörande att behålla LSTM-tillståndet för att få de önskaderesultaten. Dessutom är det inte nödvändigt att använda LSTM för enkla tidsserier.
7
van, der Westhuizen Jos. "Biological applications, visualizations, and extensions of the long short-term memory network." Thesis, University of Cambridge, 2018. https://www.repository.cam.ac.uk/handle/1810/287476.
Full textAbstract:
Sequences are ubiquitous in the domain of biology. One of the current best machine learning techniques for analysing sequences is the long short-term memory (LSTM) network. Owing to significant barriers to adoption in biology, focussed efforts are required to realize the use of LSTMs in practice. Thus, the aim of this work is to improve the state of LSTMs for biology, and we focus on biological tasks pertaining to physiological signals, peripheral neural signals, and molecules. This goal drives the three subplots in this thesis: biological applications, visualizations, and extensions. We start by demonstrating the utility of LSTMs for biological applications. On two new physiological-signal datasets, LSTMs were found to outperform hidden Markov models. LSTM-based models, implemented by other researchers, also constituted the majority of the best performing approaches on publicly available medical datasets. However, even if these models achieve the best performance on such datasets, their adoption will be limited if they fail to indicate when they are likely mistaken. Thus, we demonstrate on medical data that it is straightforward to use LSTMs in a Bayesian framework via dropout, providing model predictions with corresponding uncertainty estimates. Another dataset used to show the utility of LSTMs is a novel collection of peripheral neural signals. Manual labelling of this dataset is prohibitively expensive, and as a remedy, we propose a sequence-to-sequence model regularized by Wasserstein adversarial networks. The results indicate that the proposed model is able to infer which actions a subject performed based on its peripheral neural signals with reasonable accuracy. As these LSTMs achieve state-of-the-art performance on many biological datasets, one of the main concerns for their practical adoption is their interpretability. We explore various visualization techniques for LSTMs applied to continuous-valued medical time series and find that learning a mask to optimally delete information in the input provides useful interpretations. Furthermore, we find that the input features looked for by the LSTM align well with medical theory. For many applications, extensions of the LSTM can provide enhanced suitability. One such application is drug discovery -- another important aspect of biology. Deep learning can aid drug discovery by means of generative models, but they often produce invalid molecules due to their complex discrete structures. As a solution, we propose a version of active learning that leverages the sequential nature of the LSTM along with its Bayesian capabilities. This approach enables efficient learning of the grammar that governs the generation of discrete-valued sequences such as molecules. Efficiency is achieved by reducing the search space from one over sequences to one over the set of possible elements at each time step -- a much smaller space. Having demonstrated the suitability of LSTMs for biological applications, we seek a hardware efficient implementation. Given the success of the gated recurrent unit (GRU), which has two gates, a natural question is whether any of the LSTM gates are redundant. Research has shown that the forget gate is one of the most important gates in the LSTM. Hence, we propose a forget-gate-only version of the LSTM -- the JANET -- which outperforms both the LSTM and some of the best contemporary models on benchmark datasets, while also reducing computational cost.
8
Hernandez, Villapol Jorge Luis. "Spectrum Analysis and Prediction Using Long Short Term Memory Neural Networks and Cognitive Radios." Thesis, University of North Texas, 2017. https://digital.library.unt.edu/ark:/67531/metadc1062877/.
Full textAbstract:
One statement that we can make with absolute certainty in our current time is that wireless communication is now the standard and the de-facto type of communication. Cognitive radios are able to interpret the frequency spectrum and adapt. The aim of this work is to be able to predict whether a frequency channel is going to be busy or free in a specific time located in the future. To do this, the problem is modeled as a time series problem where each usage of a channel is treated as a sequence of busy and free slots in a fixed time frame. For this time series problem, the method being implemented is one of the latest, state-of-the-art, technique in machine learning for time series and sequence prediction: long short-term memory neural networks, or LSTMs.
9
Corni, Gabriele. "A study on the applicability of Long Short-Term Memory networks to industrial OCR." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2018.
Find full textAbstract:
This thesis summarises the research-oriented study of applicability of Long Short-Term Memory Recurrent Neural Networks (LSTMs) to industrial Optical Character Recognition (OCR) problems.
Traditionally solved through Convolutional Neural Network-based approaches (CNNs), the reported work aims to detect the OCR aspects that could be improved by exploiting recurrent patterns among pixel intensities, and speed up the overall character detection process.
Accuracy, speed and complexity act as the main key performance indicators.
After studying the core Deep Learning foundations, the best training technique to fit this problem first, and the best parametrisation next, have been selected. A set of tests eventually validated the preciseness of this solution.
The final results highlight how difficult is to perform better than CNNs for what OCR tasks are concerned. Nonetheless, with favourable background conditions, the proposed LSTM-based approach is capable of reaching a comparable accuracy rate in (potentially) less time.
10
von, Hacht Johan. "Anomaly Detection for Root Cause Analysis in System Logs using Long Short-Term Memory." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-301656.
Full textAbstract:
Many software systems are under test to ensure that they function as expected. Sometimes, a test can fail, and in that case, it is essential to understand the cause of the failure. However, as systems grow larger and become more complex, this task can become non-trivial and potentially take much time. Therefore, even partially, automating the process of root cause analysis can save time for the developers involved. This thesis investigates the use of a Long Short-Term Memory (LSTM) anomaly detector in system logs for root cause analysis. The implementation is evaluated in a quantitative and a qualitative experiment. The quantitative experiment evaluates the performance of the anomaly detector in terms of precision, recall, and F1 measure. Anomaly injection is used to measure these metrics since there are no labels in the data. Additionally, the LSTM is compared with a baseline model. The qualitative experiment evaluates how effective the anomaly detector could be for root cause analysis of the test failures. This was evaluated in interviews with an expert in the software system that produced the log data that the thesis uses. The results show that the LSTM anomaly detector achieved a higher F1 measure than the proposed baseline implementation thanks to its ability to detect unusual events and events happening out of order. The qualitative results indicate that the anomaly detector could be used for root cause analysis. In many of the evaluated test failures, the expert being interviewed could deduce the cause of the failure. Even if the detector did not find the exact issue, a particular part of the software might be highlighted, meaning that it produces many anomalous log messages. With this information, the expert could contact the people responsible for that part of the application for help. In conclusion, the anomaly detector automatically collects the necessary information for the expert to perform root cause analysis. As a result, it could save the expert time to perform this task. With further improvements, it could also be possible for non-experts to utilise the anomaly detector, reducing the need for an expert.<br>Många mjukvarusystem testas för att försäkra att de fungerar som de ska. Ibland kan ett test misslyckas och i detta fall är det viktigt att förstå varför det gick fel. Detta kan bli problematiskt när mjukvarusystemen växer och blir mer komplexa eftersom att denna uppgift kan bli icke trivial och ta mycket tid. Om man skulle kunna automatisera felsökningsprocessen skulle det kunna spara mycket tid för de invloverade utvecklarna. Denna rapport undersöker användningen av en Long Short-Term Memory (LSTM) anomalidetektor för grundorsaksanalys i loggar. Implementationen utvärderas genom en kvantitativ och kvalitativ undersökning. Den kvantitativa undersökningen utvärderar prestandan av anomalidetektorn med precision, recall och F1 mått. Artificiellt insatta anomalier används för att kunna beräkna dessa mått eftersom att det inte finns etiketter i den använda datan. Implementationen jämförs också med en annan simpel anomalidetektor. Den kvalitativa undersökning utvärderar hur användbar anomalidetektorn är för grundorsaksanalys för misslyckade tester. Detta utvärderades genom intervjuer med en expert inom mjukvaran som producerade datan som användes in denna rapport. Resultaten visar att LSTM anomalidetektorn lyckades nå ett högre F1 mått jämfört med den simpla modellen. Detta tack vare att den kunde upptäcka ovanliga loggmeddelanden och loggmeddelanden som skedde i fel ordning. De kvalitativa resultaten pekar på att anomalidetektorn kan användas för grundorsaksanalys för misslyckade tester. I många av de misslyckade tester som utvärderades kunde experten hitta anledningen till att felet misslyckades genom det som hittades av anomalidetektorn. Även om detektorn inte hittade den exakta orsaken till att testet misslyckades så kan den belysa en vissa del av mjukvaran. Detta betyder att just den delen av mjukvaran producerad många anomalier i loggarna. Med denna information kan experten kontakta andra personer som känner till den delen av mjukvaran bättre för hjälp. Anomalidetektorn automatiskt den information som är viktig för att experten ska kunna utföra grundorsaksanalys. Tack vare detta kan experten spendera mindre tid på denna uppgift. Med vissa förbättringar skulle det också kunna vara möjligt för mindre erfarna utvecklare att använda anomalidetektorn. Detta minskar behovet för en expert.
11
Svanberg, John. "Anomaly detection for non-recurring traffic congestions using Long short-term memory networks (LSTMs)." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-234465.
Full textAbstract:
In this master thesis, we implement a two-step anomaly detection mechanism for non-recurrent traffic congestions with data collected from public transport buses in Stockholm. We investigate the use of machine learning to model time series data with LSTMs and evaluate the results with a baseline prediction model. The anomaly detection algorithm embodies both collective and contextual expressivity, meaning it is capable of findingcollections of delayed buses and also takes the temporality of the data into account. Results show that the anomaly detection performance benefits from the lower prediction errors produced by the LSTM network. The intersection rule significantly decreases the number of false positives while maintaining the true positive rate at a sufficient level. The performance of the anomaly detection algorithm has been found to depend on the road segment it is applied to, some segments have been identified to be particularly hard whereas other have been identified to be easier than others. The performance of the best performing setup of the anomaly detection mechanism had a true positive rate of 84.3 % and a true negative rate of 96.0 %.<br>I den här masteruppsatsen implementerar vi en tvåstegsalgoritm för avvikelsedetektering för icke återkommande trafikstockningar. Data är insamlad från kollektivtrafikbussarna i Stockholm. Vi undersöker användningen av maskininlärning för att modellerna tidsseriedata med hjälp av LSTM-nätverk och evaluerar sedan dessa resultat med en grundmodell. Avvikelsedetekteringsalgoritmen inkluderar både kollektiv och kontextuell uttrycksfullhet, vilket innebär att kollektiva förseningar kan hittas och att även temporaliteten hos datan beaktas. Resultaten visar att prestandan hos avvikelsedetekteringen förbättras av mindre prediktionsfel genererade av LSTM-nätverket i jämförelse med grundmodellen. En regel för avvikelser baserad på snittet av två andra regler reducerar märkbart antalet falska positiva medan den höll kvar antalet sanna positiva på en tillräckligt hög nivå. Prestandan hos avvikelsedetekteringsalgoritmen har setts bero av vilken vägsträcka den tillämpas på, där några vägsträckor är svårare medan andra är lättare för avvikelsedetekteringen. Den bästa varianten av algoritmen hittade 84.3 % av alla avvikelser och 96.0 % av all avvikelsefri data blev markerad som normal data.
12
Chaulagain, Dewan. "Hybrid Analysis of Android Applications for Security Vetting." Bowling Green State University / OhioLINK, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1555608766287613.
Full text
13
Andersson, Joakim. "Evaluating Environmental Sensor Value Prediction using Machine Learning : Long Short-Term Memory Neural Networks for Smart Building Applications." Thesis, Mittuniversitetet, Institutionen för informationssystem och –teknologi, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-42852.
Full textAbstract:
IoT har blivit en stor producent av big data. Big data kan användas för att optimera operationer, för att kunna göra det så måste man kunna extrahera användbar information från big data. Detta kan göras med hjälp av neurala nätverk och maskininlärning, vilket kan leda till nya typer av smarta applikationer. Den här rapporten fokuserar på att besvara frågan hur bra är neurala nätverk på att förutspå sensor värden och hur pålitliga är förutsägelserna och om dom kan användas i verkliga applikationer. Sensorlådor användes för att samla data från olika rum och olika neurala nätverksmodeller baserade på LSTM nätverk användes för att förutspå framtida värden. Dessa värden jämfördes sedan med dom riktiga värdena och absoluta medelfelet och standardavvikelsen beräknades. Tiden som behövdes för att producera en förutsägelse mättes och medelvärde och standardavvikelsen beräknades även där. LSTM modellerna utvärderades utifrån deras prestanda och träffsäkerhet. Modellen som endast förutspådde ett värde hade bäst träffsäkerhet, och modellerna tappade träffsäkerheten desto längre in i framtiden dom försökte förutspå. Resultaten visar att även dom enkla modellerna som skapades i detta projekt kan med säkerhet förutspå värden och därför användas i olika applikationer där extremt bra förutsägelser inte behövs.<br>The IoT is becoming an increasing producer of big data. Big data can be used to optimize operations, realizing this depends on being able to extract useful information from big data. With the use of neural networks and machine learning this can be achieved and can enable smart applications that use this information. This thesis focuses on answering the question how good are neural networks at predicting sensor values and is the predictions reliable and useful in a real-life application? Sensory boxes were used to gather data from rooms, and several neural networks based on LSTM were used to predict the future values of the sensors. The absolute mean error of the predictions along with the standard deviation was calculated. The time needed to produce a prediction was measured as an absolute mean values with standard deviation. The LSTM models were then evaluated based on their performance and prediction accuracy. The single-step model, which only predicts the next timestep was the most accurate. The models loose accuracy when they need to predict longer periods of time. The results shows that simple models can predict the sensory values with some accuracy, while they may not be useful in areas where exact climate control is needed the models can be applicable in work areas such as schools or offices.
14
Hirsch, Dale A. "THE ROLE OF LEARNING MODALITY UPON LONG-TERM SPATIAL MEMORY." Cleveland State University / OhioLINK, 2013. http://rave.ohiolink.edu/etdc/view?acc_num=csu1367532907.
Full text
15
Fike, Lauren. "Cross-cultural normative indicators on the Wechsler Memory Scale (WMS) associate learning and visual reproduction subtests." Thesis, Rhodes University, 2008. http://hdl.handle.net/10962/d1002484.
Full textAbstract:
A comprehensive battery of commonly used neuropsychological tests, including the WMS Associate Learning and Visual Reproduction subtests, forming the focus of this study, were administered to a southern African sample (n = 33, age range 18-40). This sample composed of black South African, IsiXhosa speakers with an educational level of Grade 11 and 12, derived through DET and former DET schooling. The gender demographics were as follows; females n = 21 and males n = 12. This sample was purposefully selected based on current cross-cultural research which suggests that individuals matching these above-mentioned demographics are significantly disadvantaged when compared to available neuropsychological norms. This is due to the fact that current norms have been created in contexts with socio-cultural influences; including culture, language and quantity and quality of education distinctly dissimilar to individuals like that composed in the sample. Hence the purpose of this study was fourfold namely; 1) Describe and consider socio-cultural factors and the influence on test performance 2) Provide descriptive and preliminary normative data on this neuropsychologically underrepresented population 3) Compare test performance between age and gender through stratification of the sample and finally to 4) Evaluate the current norms of the two WMS subtests and assess their validity for black South Africans with DET and former DET schooling with comparisons to the results found in the study. Information derived from the statistical analyses indicated that a higher performance in favour of the younger group over the older age range was consistently found for both WMS subtests. With regards to gender, some higher means were evident for the male population in the sample than was produced by the female group. Lastly, due to the fact that most scores derived from the sample were considerably lower when compared to the available norms, it is felt that socio-cultural factors prevalent to this population are a significant cause of lower test performance and thus warrant the development of appropriate normative indicators.
16
Abraham, Aby. "Continous Speech Recognition Using Long Term Memory Cells." Ohio University / OhioLINK, 2013. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1377777011.
Full text
17
Niese, Adam Trent. "The Relationship between visual working memory and visual long-term memory." Diss., University of Iowa, 2008. https://ir.uiowa.edu/etd/210.
Full textAbstract:
This dissertation evaluated whether Visual Working Memory (VWM) is a distinct memory system or if it is an activated state of Visual Long Term Memory (VLTM). These two positions suggest different roles for VLTM representations in the performance of VWM. If VWM representations are an activated state of VLTM representations, it seems plausible that strong VLTM representations should facilitate VWM performance. However, if the two representations are actually distinct, it seems less likely that a facilitation interaction between VLTM and VWM representations should be observed.
Five experiments were conducted in which participants learned a set of trained stimuli over two days of training. Participant performance with the trained stimuli was compared to performance with novel stimuli on a subsequent VWM change detection task to determine the plausibility of VLTM-VWM interactions.
The first and second experiments revealed a LTM facilitation effect that could not be explained by priming, but the third experiment suggested that this facilitation effect was mediated by non-visual representations. The fourth and fifth experiments parceled out the contributions of non-visual memory representations, and failed to demonstrate any evidence of VLTM-VWM performance interactions.
These results, in conjunction with other examples from the literature, all converged on the conclusion that VLTM-VWM facilitation interactions are relatively implausible. As such, it was concluded that VWM and VLTM representations are discreet.
18
Abrishami, Hedayat. "Deep Learning Based Electrocardiogram Delineation." University of Cincinnati / OhioLINK, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1563525992210273.
Full text
19
Brown, Kim Freidah. "Development of long-term memory retention processes among learning disabled and nondisabled children." Diss., The University of Arizona, 1988. http://hdl.handle.net/10150/184566.
Full textAbstract:
This study investigated the development of acquisition and long term retention processes in Learning Disabled (LD) and Non-Learning Disabled children aged 7-12. One hundred six subjects were randomly assigned to memorize either a list of unrelated words (with free recall), or a list of taxonomically related words (with recall cued by category). Each subject had a 16 word list presented in visual and auditory modes. The repeated recall paradigm alternated study and test trials, with a buffer activity between trials. The acquisition phase ended when the subject reached 100% criterion. After an interval of two weeks, each subject was given 5 additional recall tests. Acquisition results indicated significant main effects for age, group (LD, Non-LD) and list type (unrelated, categorized) on measures of trial-of-last-error and total-errors. Overall, the groups which acquired the lists most quickly were the older and Non-LD subjects, with the categorized list. There was a List x Group interaction on the trial-of-last-error. With the categorized list, only age was significant, and conversely, with the unrelated list, only group was significant. On the retention measures, there were main effects for list and group, with a List x Group interaction. The only significant age effect was with total-words on the categorized list. Over the five trials (repeated measures), there was a significant effect for trials. A consistent hypermnesia effect (increase in net recall) was predominant. Further model-based analyses (Brainerd, Kingma, Howe, & Reyna, 1988) revealed storage failure, rather than retrieval failure to be the major action in children's forgetting. Learning Disabled children had significantly more storage failure than the Non-LD children. Both groups had more storage failure on the unrelated lists. There was retrieval relearning with all groups. Results are discussed within the framework of the disintegration/redintegration theory, which pertains to the gradual weakening and redintegration of bonds that unite features to form a trace.
20
Gyawali, Sanij. "Dynamic Load Modeling from PSSE-Simulated Disturbance Data using Machine Learning." Thesis, Virginia Tech, 2020. http://hdl.handle.net/10919/100591.
Full textAbstract:
Load models have evolved from simple ZIP model to composite model that incorporates
the transient dynamics of motor loads. This research utilizes the latest trend on Machine
Learning and builds reliable and accurate composite load model. A composite load model
is a combination of static (ZIP) model paralleled with a dynamic model. The dynamic
model, recommended by Western Electricity Coordinating Council (WECC), is an induction
motor representation. In this research, a dual cage induction motor with 20 parameters
pertaining to its dynamic behavior, starting behavior, and per unit calculations is used as
a dynamic model. For machine learning algorithms, a large amount of data is required.
The required PMU field data and the corresponding system models are considered Critical
Energy Infrastructure Information (CEII) and its access is limited. The next best option for
the required amount of data is from a simulating environment like PSSE. The IEEE 118 bus
system is used as a test setup in PSSE and dynamic simulations generate the required data
samples. Each of the samples contains data on Bus Voltage, Bus Current, and Bus Frequency
with corresponding induction motor parameters as target variables. It was determined that
the Artificial Neural Network (ANN) with multivariate input to single parameter output
approach worked best. Recurrent Neural Network (RNN) is also experimented side by side
to see if an additional set of information of timestamps would help the model prediction.
Moreover, a different definition of a dynamic model with a transfer function-based load
is also studied. Here, the dynamic model is defined as a mathematical representation of
the relation between bus voltage, bus frequency, and active/reactive power flowing in the
bus. With this form of load representation, Long-Short Term Memory (LSTM), a variation
of RNN, performed better than the concurrent algorithms like Support Vector Regression
(SVR). The result of this study is a load model consisting of parameters defining the load
at load bus whose predictions are compared against simulated parameters to examine their
validity for use in contingency analysis.<br>Master of Science<br>Independent system Operators (ISO) and Distribution system operators (DSO) have a responsibility to provide uninterrupted power supply to consumers. That along with the
longing to keep operating cost minimum, engineers and planners study the system beforehand and seek to find the optimum capacity for each of the power system elements like
generators, transformers, transmission lines, etc. Then they test the overall system using
power system models, which are mathematical representation of the real components, to
verify the stability and strength of the system. However, the verification is only as good as
the system models that are used. As most of the power systems components are controlled
by the operators themselves, it is easy to develop a model from their perspective. The load
is the only component controlled by consumers. Hence, the necessity of better load models.
Several studies have been made on static load modeling and the performance is on par with
real behavior. But dynamic loading, which is a load behavior dependent on time, is rather
difficult to model. Some attempts on dynamic load modeling can be found already. Physical
component-based and mathematical transfer function based dynamic models are quite widely
used for the study. These load structures are largely accepted as a good representation of
the systems dynamic behavior. With a load structure in hand, the next task is estimating
their parameters. In this research, we tested out some new machine learning methods to
accurately estimate the parameters. Thousands of simulated data are used to train machine
learning models. After training, we validated the models on some other unseen data. This
study finally goes on to recommend better methods to load modeling.
21
Rosander, Oliver, and Jim Ahlstrand. "Email Classification with Machine Learning and Word Embeddings for Improved Customer Support." Thesis, Blekinge Tekniska Högskola, Institutionen för datalogi och datorsystemteknik, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-15946.
Full textAbstract:
Classifying emails into distinct labels can have a great impact on customer support. By using machine learning to label emails the system can set up queues containing emails of a specific category. This enables support personnel to handle request quicker and more easily by selecting a queue that match their expertise. This study aims to improve the manually defined rule based algorithm, currently implemented at a large telecom company, by using machine learning. The proposed model should have higher F1-score and classification rate. Integrating or migrating from a manually defined rule based model to a machine learning model should also reduce the administrative and maintenance work. It should also make the model more flexible. By using the frameworks, TensorFlow, Scikit-learn and Gensim, the authors conduct five experiments to test the performance of several common machine learning algorithms, text-representations, word embeddings and how they work together. In this article a web based interface were implemented which can classify emails into 33 different labels with 0.91 F1-score using a Long Short Term Memory network. The authors conclude that Long Short Term Memory networks outperform other non-sequential models such as Support Vector Machines and ADABoost when predicting labels for emails.
22
Ankaräng, Fredrik, and Fabian Waldner. "Evaluating Random Forest and a Long Short-Term Memory in Classifying a Given Sentence as a Question or Non-Question." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-262209.
Full textAbstract:
Natural language processing and text classification are topics of much discussion among researchers of machine learning. Contributions in the form of new methods and models are presented on a yearly basis. However, less focus is aimed at comparing models, especially comparing models that are less complex to state-of-the-art models. This paper compares a Random Forest with a Long-Short Term Memory neural network for the task of classifying sentences as questions or non-questions, without considering punctuation. The models were trained and optimized on chat data from a Swedish insurance company, as well as user comments data on articles from a newspaper. The results showed that the LSTM model performed better than the Random Forest. However, the difference was small and therefore Random Forest could still be a preferable alternative in some use cases due to its simplicity and its ability to handle noisy data. The models’ performances were not dramatically improved after hyper parameter optimization. A literature study was also conducted aimed at exploring how customer service can be automated using a chatbot and what features and functionality should be prioritized by management during such an implementation. The findings of the study showed that a data driven design should be used, where features are derived based on the specific needs and customers of the organization. However, three features were general enough to be presented the personality of the bot, its trustworthiness and in what stage of the value chain the chatbot is implemented.<br>Språkteknologi och textklassificering är vetenskapliga områden som tillägnats mycket uppmärksamhet av forskare inom maskininlärning. Nya metoder och modeller presenteras årligen, men mindre fokus riktas på att jämföra modeller av olika karaktär. Den här uppsatsen jämför Random Forest med ett Long Short-Term Memory neuralt nätverk genom att undersöka hur väl modellerna klassificerar meningar som frågor eller icke-frågor, utan att ta hänsyn till skiljetecken. Modellerna tränades och optimerades på användardata från ett svenskt försäkringsbolag, samt kommentarer från nyhetsartiklar. Resultaten visade att LSTM-modellen presterade bättre än Random Forest. Skillnaden var dock liten, vilket innebär att Random Forest fortfarande kan vara ett bättre alternativ i vissa situationer tack vare dess enkelhet. Modellernas prestanda förbättrades inte avsevärt efter hyperparameteroptimering. En litteraturstudie genomfördes även med målsättning att undersöka hur arbetsuppgifter inom kundsupport kan automatiseras genom införandet av en chatbot, samt vilka funktioner som bör prioriteras av ledningen inför en sådan implementation. Resultaten av studien visade att en data-driven approach var att föredra, där funktionaliteten bestämdes av användarnas och organisationens specifika behov. Tre funktioner var dock tillräckligt generella för att presenteras personligheten av chatboten, dess trovärdighet och i vilket steg av värdekedjan den implementeras.
23
Bruton, Laurie. "A study of memory, learning, and emotion /." La Verne, Calif. : University of La Verne, 2003. http://0-wwwlib.umi.com.garfield.ulv.edu/dissertations/fullcit/3100047.
Full text
24
Morin, Christopher. "Learning and Long-Term Memory Formation in Danio rerio Through Two Sensory Modalities." NSUWorks, 2012. http://nsuworks.nova.edu/occ_stuetd/175.
Full textAbstract:
The zebrafish (Danio rerio) promises to meet the growing needs of gerontological and neurobehavioral research by possessing highly conserved anatomy and physiology with all other vertebrates, while having low maintenance costs and requiring only simple care. The neurological and physiological bases of learning, memory formation, and memory retention have been studied in a variety of model organisms, such as the rat, mouse, sheep, and several teleost fishes, the notable example being the zebrafish. Unfortunately, most of these animals are poorly suited to senescence research due to costs, care requirements, or long life spans. My research expands upon our rapidly growing understanding of zebrafish neurobiology, learning processes, sensory modalities, and memory retention.
Two pairs of distinct aversive conditioning experiments using classic shuttlebox designs compared the effects of sensory modality and conditioned-unconditioned stimulus (CS-US) intervals, the time delay between application of conditioned sensory stimulus and delivery of the stressful unconditioned stimulus in the event of failure to avoid it, upon memory formation and retention. These studies yielded a general spectrum of results against which future conditioning studies may be compared. Both visual and olfactory stimuli were tested, as were 10 second and 15 second CS-US intervals. Successes were scored when the fish crossed the shuttlebox hurtle within the CS-US interval, thereby avoiding the negative unconditioned stimulus. After a three-month delay, ten additional trials were conducted to compare the long-term memory retention resulting from each protocol.
When testing a 15s CS-US interval, olfactory conditioning was significantly more likely (39%) to produce a successful outcome (memory formation) than visual conditioning. Grouped results reveal that the second pair of experiments, each with a 10s CS-US interval, yielded significantly more successful memory formation than a 15s CS-US interval. A significant difference was found when comparing the results of any two experiments, except between the results of the visual and olfactory 10s interval experiments). Only the olfactory experiment using a 15s CS-US interval yielded memory retention results significantly higher than the mean of memory retention results from the four experiments.
These findings offer inconclusive evidence supporting olfaction’s strong role in memory formation and retention in zebrafish. The results expand our understanding of the relationship between the olfactory and visual senses and memory in the zebrafish and indicate the olfactory sense’s key role in vertebrate neurobiology, warranting further research into the effects of aging on the olfactory-memory modality.
25
Beck, Whitney Marie. "Describing time spent using various teaching techniques and student immediate, short-term, and long-term cognitive retention." Columbus, Ohio : Ohio State University, 2009. http://rave.ohiolink.edu/etdc/view?acc%5Fnum=osu1243523704.
Full text
26
Capshaw, Riley. "Relation Classification using Semantically-Enhanced Syntactic Dependency Paths : Combining Semantic and Syntactic Dependencies for Relation Classification using Long Short-Term Memory Networks." Thesis, Linköpings universitet, Interaktiva och kognitiva system, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-153877.
Full textAbstract:
Many approaches to solving tasks in the field of Natural Language Processing (NLP) use syntactic dependency trees (SDTs) as a feature to represent the latent nonlinear structure within sentences. Recently, work in parsing sentences to graph-based structures which encode semantic relationships between words—called semantic dependency graphs (SDGs)—has gained interest. This thesis seeks to explore the use of SDGs in place of and alongside SDTs within a relation classification system based on long short-term memory (LSTM) neural networks. Two methods for handling the information in these graphs are presented and compared between two SDG formalisms. Three new relation extraction system architectures have been created based on these methods and are compared to a recent state-of-the-art LSTM-based system, showing comparable results when semantic dependencies are used to enhance syntactic dependencies, but with significantly fewer training parameters.
27
Gudmundsson, Johan, and Francis Menkes. "Swedish Natural Language Processing with Long Short-term Memory Neural Networks : A Machine Learning-powered Grammar and Spell-checker for the Swedish Language." Thesis, Linnéuniversitetet, Institutionen för datavetenskap och medieteknik (DM), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-76819.
Full textAbstract:
Natural Language Processing (NLP) is a field studying computer processing of human language. Recently, neural network language models, a subset of machine learning, have been used to great effect in this field. However, research remains focused on the English language, with few implementations in other languages of the world. This work focuses on how NLP techniques can be used for the task of grammar and spelling correction in the Swedish language, in order to investigate how language models can be applied to non-English languages. We use a controlled experiment to find the hyperparameters most suitable for grammar and spelling correction on the Göteborgs-Posten corpus, using a Long Short-term Memory Recurrent Neural Network. We present promising results for Swedish-specific grammar correction tasks using this kind of neural network; specifically, our network has a high accuracy in completing these tasks, though the accuracy achieved for language-independent typos remains low.
28
Berry, Chadwick Alan. "The fidelity of long-term memory for perceptual magnitudes, symbolic vs. metric learning paradigms." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 1997. http://www.collectionscanada.ca/obj/s4/f2/dsk2/ftp01/MQ29183.pdf.
Full text
29
Mishra, Vishal Vijayshankar. "Sequence-to-Sequence Learning using Deep Learning for Optical Character Recognition (OCR)." University of Toledo / OhioLINK, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=toledo1513273051760905.
Full text
30
Bilgin, Zikri. "Long-term Potentiation In Teaching Vocabulary In Foreign Language." Master's thesis, METU, 2010. http://etd.lib.metu.edu.tr/upload/12611684/index.pdf.
Full textAbstract:
This thesis mainly intends to study and reach some conclusions related to major challenges concerning vocabulary teaching or learning, how vocabulary teaching can be improved, findings obtained from the studies in order to reach that purpose and to what extend the suggested alternative vocabulary techniques are effective. It is also aimed to outline the basic insights of the mind, storage, and retrieval from the literature involving linguistics and language teaching. Based on above mentioned background knowledge, it is also intended to derive some significant conclusions to improve the effectiveness and thus the quality of vocabulary teaching in language instruction. In accordance with the principles of the human memory, how we can alter current vocabulary instruction techniques and activities and what scholars offer language teachers and learners are dealt with in detail. So as to validate and prove the efficiency of suggested techniques and activities, a case study is carried out and findings are discussed at large. Additionally, interviews about vocabulary teaching have been carried out with the involved students and instructors and the obtained data has been evaluated. In the final part of the research, some implications and suggestion related to vocabulary teaching are provided along with the underlying rationale behind them aiming to increase the quality of teaching of lexical items and as a result to increase overall quality of language instruction.
31
Tovedal, Sofiea. "On The Effectiveness of Multi-TaskLearningAn evaluation of Multi-Task Learning techniques in deep learning models." Thesis, Umeå universitet, Institutionen för datavetenskap, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-172257.
Full textAbstract:
Multi-Task Learning is today an interesting and promising field which many mention as a must for achieving the next level advancement within machine learning. However, in reality, Multi-Task Learning is much more rarely used in real-world implementations than its more popular cousin Transfer Learning. The questionis why that is and if Multi-Task Learning outperforms its Single-Task counterparts. In this thesis different Multi-Task Learning architectures were utilized in order to build a model that can handle labeling real technical issues within two categories. The model faces a challenging imbalanced data set with many labels to choose from and short texts to base its predictions on. Can task-sharing be the answer to these problems? This thesis investigated three Multi-Task Learning architectures and compared their performance to a Single-Task model. An authentic data set and two labeling tasks was used in training the models with the method of supervised learning. The four model architectures; Single-Task, Multi-Task, Cross-Stitched and the Shared-Private, first went through a hyper parameter tuning process using one of the two layer options LSTM and GRU. They were then boosted by auxiliary tasks and finally evaluated against each other.
32
Jonsson, Max. "Deep Learning för klassificering av kundsupport-ärenden." Thesis, Högskolan i Gävle, Datavetenskap, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:hig:diva-32687.
Full textAbstract:
Företag och organisationer som tillhandahåller kundsupport via e-post kommer över tid att samla på sig stora mängder textuella data. Tack vare kontinuerliga framsteg inom Machine Learning ökar ständigt möjligheterna att dra nytta av tidigare insamlat data för att effektivisera organisationens framtida supporthantering. Syftet med denna studie är att analysera och utvärdera hur Deep Learning kan användas för att automatisera processen att klassificera supportärenden. Studien baseras på ett svenskt företags domän där klassificeringarna sker inom företagets fördefinierade kategorier. För att bygga upp ett dataset extraherades supportärenden inkomna via e-post (par av rubrik och meddelande) från företagets supportdatabas, där samtliga ärenden tillhörde en av nio distinkta kategorier. Utvärderingen gjordes genom att analysera skillnaderna i systemets uppmätta precision då olika metoder för datastädning användes, samt då de neurala nätverken byggdes upp med olika arkitekturer. En avgränsning gjordes att endast undersöka olika typer av Convolutional Neural Networks (CNN) samt Recurrent Neural Networks (RNN) i form av både enkel- och dubbelriktade Long Short Time Memory (LSTM) celler. Resultaten från denna studie visar ingen ökning i precision för någon av de undersökta datastädningsmetoderna. Dock visar resultaten att en begränsning av den använda ordlistan heller inte genererar någon negativ effekt. En begränsning av ordlistan kan fortfarande vara användbar för att minimera andra effekter så som exempelvis träningstiden, och eventuellt även minska risken för överanpassning. Av de undersökta nätverksarkitekturerna presterade CNN bättre än RNN på det använda datasetet. Den mest gynnsamma nätverksarkitekturen var ett nätverk med en konvolution per pipeline som för två olika test-set genererade precisioner på 79,3 respektive 75,4 procent. Resultaten visar också att några kategorier är svårare för nätverket att klassificera än andra, eftersom dessa inte är tillräckligt distinkta från resterande kategorier i datasetet.<br>Companies and organizations providing customer support via email will over time grow a big corpus of text documents. With advances made in Machine Learning the possibilities to use this data to improve the customer support efficiency is steadily increasing. The aim of this study is to analyze and evaluate the use of Deep Learning methods for automizing the process of classifying support errands. This study is based on a Swedish company’s domain where the classification was made within the company’s predefined categories. A dataset was built by obtaining email support errands (subject and body pairs) from the company’s support database. The dataset consisted of data belonging to one of nine separate categories. The evaluation was done by analyzing the alteration in classification accuracy when using different methods for data cleaning and by using different network architectures. A delimitation was set to only examine the effects by using different combinations of Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) in the shape of both unidirectional and bidirectional Long Short Time Memory (LSTM) cells. The results of this study show no increase in classification accuracy by any of the examined data cleaning methods. However, a feature reduction of the used vocabulary is proven to neither have any negative impact on the accuracy. A feature reduction might still be beneficial to minimize other side effects such as the time required to train a network, and possibly to help prevent overfitting. Among the examined network architectures CNN were proven to outperform RNN on the used dataset. The most accurate network architecture was a single convolutional network which on two different test sets reached classification rates of 79,3 and 75,4 percent respectively. The results also show some categories to be harder to classify than others, due to them not being distinct enough towards the rest of the categories in the dataset.
33
Chari, Shreya Krishnama. "Link blockage modelling for channel state prediction in high-frequencies using deep learning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-287458.
Full textAbstract:
With the accessibility to generous spectrum and development of high gain antenna arrays, wireless communication in higher frequency bands providing multi-gigabit short range wireless access has become a reality. The directional antennas have proven to reduce losses due to interfering signals but are still exposed to blockage events. These events impede the overall user connectivity and throughput. A mobile blocker such as a moving vehicle amplifies the blockage effect. Modelling the blockage effects helps in understanding these events in depth and in maintaining the user connectivity. This thesis proposes the use of a four state channel model to describe blockage events in high-frequency communication. Two deep learning architectures are then designed and evaluated for two possible tasks, the prediction of the signal strength and the classification of the channel state. The evaluations based on simulated traces show high accuracy, and suggest that the proposed models have the potential to be extended for deployment in real systems.<br>Med tillgängligheten till generöst spektrum och utveckling av antennmatriser med hög förstärkning har trådlös kommunikation i högre frekvensband som ger multi-gigabit kortdistans trådlös åtkomst blivit verklighet. Riktningsantennerna har visat sig minska förluster på grund av störande signaler men är fortfarande utsatta för blockeringshändelser. Dessa händelser hindrar den övergripande användaranslutningen och genomströmningen. En mobil blockerare såsom ett fordon i rörelse förstärker blockeringseffekten. Modellering av blockeringseffekter hjälper till att förstå dessa händelser på djupet och bibehålla användaranslutningen. Denna avhandling föreslår användning av en fyrstatskanalmodell för att beskriva blockeringshändelser i högfrekvent kommunikation. Två djupinlärningsarkitekturer designas och utvärderas för två möjliga uppgifter, förutsägelsen av signalstyrkan och klassificeringen av kanalstatusen. Utvärderingarna baserade på simulerade spår visar hög noggrannhet och föreslår att de föreslagna modellerna har potential att utökas för distribution i verkliga system.
34
Komol, Md Mostafizur Rahman. "C-ITS based prediction of driver red light running and turning behaviours." Thesis, Queensland University of Technology, 2022. https://eprints.qut.edu.au/227694/1/Md%20Mostafizur%20Rahman_Komol_Thesis.pdf.
Full textAbstract:
Red light running is a major traffic violation. Drivers often aggressively or unintentionally violate red signal and cause traffic collisions. Moreover, Vision impairment of turning vehicles by large vehicles and road side static structures near intersections often lead to VRU crashes during their crossing at the intersection. In this research, we have developed models to predict drivers’ red light running and turning behaviour at intersections using Long Short Term Memory and Gated Recurrent Unit algorithms. We have used vehicle kinematic dataset of the C-ITS project: Ipswich Connected Vehicle Pilot, Queensland, taken from the Department of Transport and Main Road, Queensland.
35
Gulbrandsen-MacDonald, Tine L., and University of Lethbridge Faculty of Arts and Science. "The role of the hippocampus and post-learning hippocampal activity in long-term consolidation of context memory." Thesis, Lethbridge, Alta. : University of Lethbridge, Dept. of Neuroscience, c2011, 2011. http://hdl.handle.net/10133/2635.
Full textAbstract:
Sutherland, Sparks and Lehmann (2010) proposed a new theory of memory consolidation, termed Distributed Reinstatement Theory (DRT), where the hippocampus (HPC) is needed for initial encoding but some types of memories are established in non-HPC systems through post-learning HPC activity. An evaluation of the current methodology of temporary inactivation was conducted experimentally. By permanently implanting two bilateral guide cannulae in the HPC and infusing ropivacaine cellular activity could be reduced by 97%. Rats were trained in a context-fear paradigm. Six learning episodes distributed across three days made the memory resistant to HPC inactivation while three episodes did not. Blocking post-learning HPC activity following three of six training sessions failed to reduce the rat’s memory of the fearful context. These results fail to support DRT and indicate that one or more memory systems outside the HPC can acquire context memory without HPC post-event activity.<br>x, 85 leaves : ill. ; 29 cm
36
Sparks, Fraser T. "Interactions of the hippocampus and non-hippocampal long-term memory systems during learning, remembering, and over time." Thesis, Lethbridge, Alta. : University of Lethbridge, Dept. of Neuroscience, c2012, 2012. http://hdl.handle.net/10133/3116.
Full textAbstract:
The hippocampus and non-hippocampal long-term memory systems each have the capacity
to learn and express contextual fear memory. How these systems interact during learning
and remembering revolves around hippocampal mediated interference, where the hippocampus
dominates for both the acquisition and expression of long-term memory. Hippocampal
interference during learning can be overcome by modifying learning parameters
such that learning is distributed across multiple independent sessions. The standard view of
the role of the hippocampus in long-term memory retrieval is that it is temporally limited,
where recently acquired memory is dependent on hippocampal function though as a memory
ages, dependency is transferred to other memory systems by a process called systems
consolidation. Distributed training demonstrates that learning parameters create a memory
that is resistant to hippocampal damage. We find little evidence to support temporally based
systems consolidation, and present data that supports the view that if the hippocampus is
initially involved in learning a memory, it will always be necessary for accurate retrieval
of that memory. A critical assessment of the rat literature revealed that initial memory
strength, and/or lesion techniques might be responsible for the few studies that report temporally
graded retrograde amnesia using contextual fear conditioning. Our experiments
designed to directly test these possibilities resulted in flat gradients, providing further evidence
that the hippocampus plays a permanent role in long-term memory retrieval. We
propose and assess alternatives to the standard model and conclude that a dual store model
is most parsimonious within the presented experiments and related literature. Interactions
of the hippocampus and non-hippocampal systems take place at the time of learning and
remembering, and are persistent over time.<br>xvi, 161 leaves : ill. (some col.) ; 29 cm
37
Fors, Johansson Christoffer. "Arrival Time Predictions for Buses using Recurrent Neural Networks." Thesis, Linköpings universitet, Artificiell intelligens och integrerade datorsystem, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-165133.
Full textAbstract:
In this thesis, two different types of bus passengers are identified. These two types, namely current passengers and passengers-to-be have different needs in terms of arrival time predictions. A set of machine learning models based on recurrent neural networks and long short-term memory units were developed to meet these needs. Furthermore, bus data from the public transport in Östergötland county, Sweden, were collected and used for training new machine learning models. These new models are compared with the current prediction system that is used today to provide passengers with arrival time information. The models proposed in this thesis uses a sequence of time steps as input and the observed arrival time as output. Each input time step contains information about the current state such as the time of arrival, the departure time from thevery first stop and the current position in Cartesian coordinates. The targeted value for each input is the arrival time at the next time step. To predict the rest of the trip, the prediction for the next step is simply used as input in the next time step. The result shows that the proposed models can improve the mean absolute error per stop between 7.2% to 40.9% compared to the system used today on all eight routes tested. Furthermore, the choice of loss function introduces models thatcan meet the identified passengers need by trading average prediction accuracy for a certainty that predictions do not overestimate or underestimate the target time in approximately 95% of the cases.
38
Bajic, Daniel Andrew. "The temporal dynamics of strategy execution in cognitive skill learning." Diss., [La Jolla] : University of California, San Diego, 2009. http://wwwlib.umi.com/cr/ucsd/fullcit?p3369155.
Full textAbstract:
Thesis (Ph. D.)--University of California, San Diego, 2009.<br>Title from first page of PDF file (viewed September 15, 2009). Available via ProQuest Digital Dissertations. Vita. Includes bibliographical references.
39
Gryzelius, Thomas. "El aprendizaje distribuido como estrategia didáctica en la enseñanza del vocabulario de ELE : Un acercamiento a su uso en el salón escolar sueco." Thesis, Karlstads universitet, Institutionen för språk, litteratur och interkultur, 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-42505.
Full textAbstract:
Aprender nuevas palabras en un idioma extranjero, es decir, el léxico necesario que fundamenta la posibilidad del desarrollo de las destrezas comunicativas, constituye uno de los problemas más complejos en el proceso tanto de enseñanza como de aprendizaje del español como lengua extranjera. En relación con el aprendizaje del vocabulario identificamos un posible problema; el riesgo de que el número de palabras aprendidas se olvide aumenta después de la prueba o los ejercicios. Si nuestros alumnos no pueden ampliar su vocabulario su competencia comunicativa tampoco va a desarrollar.Para poder entender por qué ocurre el problema y cómo se podría encontrar otros recursos didácticos que contribuyan a un cambio en el proceso, investigamos un fenómeno conocido por la psicología de la educación como el efecto de la memoria espaciada - un fenómeno cognitivo que se benéfica de las repeticiones, pero siempre distribuidas en el tiempo. Estrategias de enseñanza que utilizan dicho efecto se refiere como aprendizaje distribuido.Mediante un pequeño estudio analizamos el efecto de la memoria espaciada (ME) como método alternativa. De este estudio podemos inferir que existe un efecto de memoria espaciada tangible en el aprendizaje de los alumnos que estudiaron según un modelo distribuido, es decir con repeticiones.Pudimos constatar un resultado positivo en este pequeño estudio piloto. Los alumnos lograron recordar en la examinación el 85% de las palabras ejercitadas en la clase un mes después. Este resultado abre nuevas perspectivas de estudio e indica que puede haber alternativas didácticas en la enseñanza del vocabulario de ELE en el salón escolar sueco.<br>Studying and learning words in a foreign language in order to develop a vocabulary that promotes communicative competence, is a daunting task both for students and teachers of Spanish as a foreign language in Sweden and elsewhere. In this context we identify one problem; the possibility that words learned in class will be forgotten as soon as they have been tested on a quiz or exam. If our students cannot incorporate new words into their vocabulary it is quite possible that their communicative development will stop or slow down. In order to understand this problem and find alternative ways to teach vocabulary we investigated a phenomena called ‘the spaced memory effect’. In the field of educational psychology this is when the learner study with repetitions distributed over time. Practices that build on the spaced memory effect are often called distributed learning. In a small study we tested this effect as an alternative way of teaching and learning vocabulary. From this study we could conclude that the effect is possible to measure and that it is consistent in all the test subjects that followed a study model that was based on repetitions or a distributed learning model. It was shown that after one month the students were able to remember 85% of the words. The results from this small study provides new perspectives for further investigation and suggests that there are alternative ways of teaching Spanish vocabulary in Swedish schools.
40
Bertani, Federico. "Deep Learning methods for Portfolio Optimization." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2021. http://amslaurea.unibo.it/24245/.
Full textAbstract:
Portfolio optimization is one of the most studied fields that have been researched with machine learning approaches because of its inherent demand for forecasting future market properties. In this thesis, it is shown how one can use deep neural networks with historical returns to do risk adjusted asset allocation. Unlike previous studies which set as target variable asset prices, the variable to predict here is represented by the best asset allocation strategy. Experiments performed on a time period of seven years show that temporal convolutional networks are superior to long short term memory networks and transformers. Compared to baseline benchmarks, the computed allocation has an average increase in the year revenue between 2% and 5%. Furthermore, results are compared against equally weighted, inverse volatility and risk parity methods, showing higher cumulative returns than the first method and equal if not higher in some cases than the latter methods.
41
Fatah, Kiar, and Taariq Nazar. "Stock Market Prediction With Deep Learning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-293853.
Full textAbstract:
Due to the unpredictability of the stock market,forecasting stock prices is a challenging task. In this project,we will investigate the performance of the machine learningalgorithm LSTM for stock market prediction. The algorithmwill be based only on historical numerical data and technicalindicators for IBM and FORD. Furthermore, the denoising anddimension reduction algorithm, PCA, is applied to the stockdata, to examine if the performance of forecasting the stockprice is greater than the initial model. A second method, transferlearning, is applied by training the model on the IBM datasetand then applying it on the FORD dataset, and vice versa, toevaluate if the results will improve. The results show that whenthe PCA algorithm is applied to the dataset separately, and incombination with transfer learning, the performance is greater incomparison to the initial model. Moreover, the transfer learningmodel is inconsistent as the performance is worse for FORD inrespect to the initial model, but better for IBM. Thus, concerningthe results when forecasting stock prices using related tools, it issuggested to use trial and error to identify which of the modelsthat performs the optimally.<br>Att förutse aktiekurser är en utmanande uppgift. Detta beror på aktiemarknadens oförutsägbarhet. Därför kommer vi i detta projekt att undersöka prestandan för maskininlärnings algoritmen LSTMs prognosförmåga för aktie priser. Algoritmen baseras endast på historisk numerisk data och tekniska indikatorer for företagen IBM och FORD. Vidare tillämpas brus minskande och dimension reducerande algorithmen, PCA, på aktiedata för att undersöka om prestandan för att förutse aktie priser är bättre än den ursprungliga modellen. En andra metod, transfer learning, tillämpas genom att träna modellen på IBM data och sedan använda den på FORD data, och vice versa, för att utvärdera om resultaten kommer att förbättras. Resultaten visar, när PCA-algoritmen tillämpas på aktiedata separat, och i kombination med transfer learning är prestandan bättre jämfört med bas modellen. Vidare kan vi inte dra slutsatser om transfer learning då prestandan är sämre för FORD med avseende på bas modellen, men bättre för IBM. I hänsyn till resultaten så föreslås det att man tillämpar modellerna för att identifiera vilken som är mest optimal när man arbetar i ett relaterat ämnesområde.<br>Kandidatexjobb i elektroteknik 2020, KTH, Stockholm
42
Bergelin, Victor. "Human Activity Recognition and Behavioral Prediction using Wearable Sensors and Deep Learning." Thesis, Linköpings universitet, Matematiska institutionen, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-138064.
Full textAbstract:
When moving into a more connected world together with machines, a mutual understanding will be very important. With the increased availability in wear- able sensors, a better understanding of human needs is suggested. The Dart- mouth Research study at the Psychiatric Research Center has examined the viability of detecting and further on predicting human behaviour and complex tasks. The field of smoking detection was challenged by using the Q-sensor by Affectiva as a prototype. Further more, this study implemented a framework for future research on the basis for developing a low cost, connected, device with Thayer Engineering School at Dartmouth College. With 3 days of data from 10 subjects smoking sessions was detected with just under 90% accuracy using the Conditional Random Field algorithm. However, predicting smoking with Electrodermal Momentary Assessment (EMA) remains an unanswered ques- tion. Hopefully a tool has been provided as a platform for better understanding of habits and behaviour.
43
Fongo, Daniele. "Previsione del declino funzionale tramite l'utilizzo di Reti Neurali Ricorrenti." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2017. http://amslaurea.unibo.it/14889/.
Full textAbstract:
PreventIT è un progetto europeo nato con il fine di prevenire il declino funzionale e l'insorgere di disabilità in persone prossime all'anzianità. Questo viene fatto da una parte promuovendo un invecchiamento salutare attraverso l'uso di dispositivi tecnologici come smartphone e smartwatch che incoraggino l'attività fisica, dall'altra parte effettuando degli screening continui, osservazioni o controlli periodici per analizzare i rischi del declino ed individuare le persone più in pericolo al fine di intervenire il prima possibile.
A partire dagli stessi dati utilizzati all'interno del progetto europeo sopra citato, lo scopo della tesi è stato quello di sviluppare un tool parallelo basato su Reti Neurali Artificiali in grado di automatizzare tale analisi del rischio, offrendo una buona previsione del possibile declino funzionale futuro a partire da una serie di osservazioni sulle persone. In particolare, l'interesse scientifico del progetto è stato nel constatare quale fosse il modello di rete neurale che meglio riuscisse a predire una classe di rischio partendo da uno scenario complesso in cui le osservazioni risultano eterogenee poiché estrapolate da multipli dataset differenti.
I risultati sperimentali dimostrano che l’utilizzo di Long Short-Term Memory e di Deep Feedforward garantiscono ottime previsioni di declino funzionale, con un AUCROC pari a 0.881 e 0.883 rispettivamente. Ciò indica che un approccio ricorrente ed un’analisi temporale di intere sequenze di osservazioni potrebbero non risultare necessari per la predizione del declino funzionale.
44
SKELTON, MATTHEW RYAN. "EFFECTS OF NEONATAL 3,4-METHYLENEDIOXYMETHAMPHETAMINE ON HIPPOCAMPAL GENE EXPRESSION, SPATIAL LEARNING AND LONG-TERM POTENTIATION." University of Cincinnati / OhioLINK, 2006. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1148067008.
Full text
45
PASINI, SILVIA. "Role of activated transcription factor 4 (ATF4) in learning and memory." Doctoral thesis, Università degli Studi di Milano-Bicocca, 2011. http://hdl.handle.net/10281/27132.
Full textAbstract:
The aim of this study is to understand the role of Activated Transcription Factor 4 (ATF4) in the processes of learning and memory. The topic of learning and memory has always aroused great interest from time immemorial and although a lot of researches have been focused on this subject for a long time, many mechanisms have not yet been fully understood.
Identifying the players and the mechanisms involved in learning and memory is of utmost importance because deficits in these cognitive functions are symptoms of common neurological diseases like stoke, depression, dementia and Alzheimer’s disease, one of the most wide spread neurodegenerative disease.
It has already been established that new gene expression and protein synthesis are required for long term memory, providing the basis to think that transcription factors may play a key role in these processes. Several studies have demonstrated the involvement of different transcription factors in memory formation such as cAMP response element binding protein (CREB), CCAAT enhancer binding protein (C/EBP), activated protein 1 (AP1), early growth response factor (Egr) and Rel/nuclear factor kB (Rel/NFkB).
Very little is known about the involvement of another transcription factor, Activated Transcription Factor 4. ATF4 is a member of the activated transcription factor (ATF)/cyclic AMP response element binding protein (CREB) family. It was originally described as a repressor of CRE-dependent gene transcription but recent studies have shown it to be a transcriptional activator. It is also a stress responsive gene, regulating the adaptation of cells to stress stimuli such as anoxia, endoplasmic reticulum stress and oxidative stress. ATF4 plays an essential role in development, and is particularly required for proper skeletal and eye development and is also involved in tumor progression and metastasis.
ATF4 has always been reported as a memory repressor that blocks new gene expression required for memory formation but no study has ever investigated it in a specific and direct way. The aim of this thesis is to study, in a specific and direct manner, the role of ATF4 in the processes of learning and memory. To reach this goal, ATF4 expression was modified in mouse hippocampi, the brain region mainly involved in learning and memory, with the injection of lentivirus carrying ATF4 gene, for the gain-of-function analysis, and lentivirus carrying shATF4, for the loss-of-function studies.
Before starting the experiments of ATF4 overexpression and downregulation, preliminary experiments were conducted to set up the injection coordinates to target the mouse hippocampi, to verify the lentiviral tropism and most importantly to evaluate the lentiviral spread, within the hippocampus, after the injection.
The consequence of ATF4 gain- and loss-of-function was then studied in the performance of standard behavioral tests such as Water Maze tests and Fear Conditioning, widely used to assess spatial and associative memory respectively. The behavioral test results showed that ATF4 protein overexpression enhances spatial memory, under the weak training paradigm in the Morris Water Maze test, and associative memory while ATF4 downregulation impairs spatial memory under the standard training condition.
After completing the behavioral tests, ATF4 overexpressed and downregulated mice were subjected to electrophysiological and neuronal spine analysis to verify if the alteration in cognitive functions, as a result of ATF4 modification, is supported by changes in synaptic potentiation and spine density and morphology.
Long Term Potentiation (LTP) is a long lasting enhancement in neuronal transmission and is widely considered as a cellular model of learning and memory in the central nervous system. The long-term memory impairment of ATF4 downregulated mice is supported by electrophysiological analysis, in which ATF4 downregulated slices showed an impairment in LTP. Unexpectedly, LTP impairment was also found in ATF4 overexpressed slices, maybe due to the difference in the time between the injection and the behavioral tests or the electrophysiological recordings.
Most of the intracellular pathways responsible for LTP require new gene expression and protein synthesis. This, in turn, leads to morphological changes required to sustain the enhancement of signal transmission. One of these morphological changes is the modification of the density and the morphology of dendritic spines. ATF4 up- and downregulation in hippocampal neurons does not affect spine density but ATF4 overexpression causes a significant increase in the percentage of mushroom spines as compared to that found after ATF4 downregulation. Mushroom spines with a large head are the most stable neuronal spines and contribute to strong synaptic connections, hence it has been hypothesized that they represent the “memory spines”.
Collectively, these results support the hypothesis that the transcription factor ATF4 plays a positive role in synaptic plasticity and memory formation. Further studies need to be done to understand the molecular mechanisms through which ATF4 acts.
This thesis represents only a step on the road towards understanding the complicate mechanisms of learning and memory, not forgetting that the most important discoveries were the result of small knowledge acquired step by step.
46
Odinsdottir, Gudny Björk, and Jesper Larsson. "Deep Learning Approach for Extracting Heart Rate Variability from a Photoplethysmographic Signal." Thesis, Högskolan Kristianstad, Fakulteten för naturvetenskap, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:hkr:diva-21368.
Full textAbstract:
Photoplethysmography (PPG) is a method to detect blood volume changes in every heartbeat. The peaks in the PPG signal corresponds to the electrical impulses sent by the heart. The duration between each heartbeat varies, and these variances are better known as heart rate variability (HRV). Thus, finding peaks correctly from PPG signals provides the opportunity to measure an accurate HRV. Additional research indicates that deep learning approaches can extract HRV from a PPG signal with significantly greater accuracy compared to other traditional methods. In this study, deep learning classifiers were built to detect peaks in a noise-contaminated PPG signal and to recognize the performed activity during the data recording. The dataset used in this study is provided by the PhysioBank database consisting of synchronized PPG-, acceleration- and gyro data. The models investigated in this study were limited toa one-layer LSTM network with six varying numbers of neurons and four different window sizes. The most accurate model for the peak classification was the model consisting of 256 neurons and a window size of 15 time steps, with a Matthews correlation coefficient (MCC) of 0.74. The model consisted of64 neurons and a window duration of 1.25 seconds resulted in the most accurate activity classification, with an MCC score of 0.63. Concludingly, more optimization of a deep learning approach could lead to promising accuracy on peak detection and thus an accurate measurement of HRV. The probable cause for the low accuracy of the activity classification problem is the limited data used in this study.
47
Pavllo, Dario. "Riconoscimento real-time di gesture tramite tecniche di machine learning." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2016. http://amslaurea.unibo.it/10999/.
Full textAbstract:
Il riconoscimento delle gesture è un tema di ricerca che sta acquisendo sempre più popolarità, specialmente negli ultimi anni, grazie ai progressi tecnologici dei dispositivi embedded e dei sensori. Lo scopo di questa tesi è quello di utilizzare alcune tecniche di machine learning per realizzare un sistema in grado di riconoscere e classificare in tempo reale i gesti delle mani, a partire dai segnali mioelettrici (EMG) prodotti dai muscoli. Inoltre, per consentire il riconoscimento di movimenti spaziali complessi, verranno elaborati anche segnali di tipo inerziale, provenienti da una Inertial Measurement Unit (IMU) provvista di accelerometro, giroscopio e magnetometro.
La prima parte della tesi, oltre ad offrire una panoramica sui dispositivi wearable e sui sensori, si occuperà di analizzare alcune tecniche per la classificazione di sequenze temporali, evidenziandone vantaggi e svantaggi. In particolare, verranno considerati approcci basati su Dynamic Time Warping (DTW), Hidden Markov Models (HMM), e reti neurali ricorrenti (RNN) di tipo Long Short-Term Memory (LSTM), che rappresentano una delle ultime evoluzioni nel campo del deep learning.
La seconda parte, invece, riguarderà il progetto vero e proprio. Verrà impiegato il dispositivo wearable Myo di Thalmic Labs come caso di studio, e saranno applicate nel dettaglio le tecniche basate su DTW e HMM per progettare e realizzare un framework in grado di eseguire il riconoscimento real-time di gesture. Il capitolo finale mostrerà i risultati ottenuti (fornendo anche un confronto tra le tecniche analizzate), sia per la classificazione di gesture isolate che per il riconoscimento in tempo reale.
48
Dittinger, Eva Maria. "From auditory perception to memory : musicianship as a window into novel word learning." Thesis, Aix-Marseille, 2018. http://www.theses.fr/2018AIXM0513/document.
Full textAbstract:
Les avantages liés à la formation musicale transfèrent au traitement du langage, et à certaines fonctions perceptives et cognitives. Nous examinons si cette formation facilite aussi l'apprentissage de mots nouveaux au long de la vie. Les enfants «musiciens» et les jeunes musiciens professionnels surpassent les participants de contrôle dans une série d’expériences, avec une plasticité cérébrale plus rapide, et une connectivité fonctionnelle plus forte, mesurées par électroencéphalographie. Les résultats des musiciens plus âgés sont moins clairs, suggérant un impact limité de la formation musicale sur le déclin cognitif. Enfin, les jeunes musiciens ont une meilleure mémoire à long terme des nouveaux mots, ce qui contribuerait à expliquer l’avantage observé. Ces effets de transfert de la formation musicale au niveau sémantique et de la mémoire à long terme révèlent l’importance des fonctions cognitives générales et ouvrent de nouvelles perspectives pour l’éducation et la rééducation<br>Based on results evidencing music training-related advantages on speech processing, perceptive and cognitive functions, we examine whether music training facilitates novel word learning throughout the lifespan. We show that musically-trained children and young professional musicians outperform controls in a series of experiments, with faster brain plasticity and stronger functional connectivity, as measured by electroencephalography. By contrast, advantages for old adult musicians are less clear-cut, suggesting a limited impact of music training to counteract cognitive decline. Finally, young musicians show better long-term memory for novel words, which possibly contributes, along with better auditory perception and attention, to their advantage in word learning. By showing transfer effects from music training to semantic processing and long-term memory, results reveal the importance of domain-general cognitive functions and open new perspectives for education and rehabilitation
49
Norgren, Eric. "Pulse Repetition Interval Modulation Classification using Machine Learning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-241152.
Full textAbstract:
Radar signals are used for estimating location, speed and direction of an object. Some radars emit pulses, while others emit a continuous wave. Both types of radars emit signals according to some pattern; a pulse radar, for example, emits pulses with a specific time interval between pulses. This time interval may either be stable, change linearly, or follow some other pattern. The interval between two emitted pulses is often referred to as the pulse repetition interval (PRI), and the pattern that defines the PRI is often referred to as the modulation. Classifying which PRI modulation is used in a radar signal is a crucial component for the task of identifying who is emitting the signal. Incorrectly classifying the used modulation can lead to an incorrect guess of the identity of the agent emitting the signal, and can as a consequence be fatal. This work investigates how a long short-term memory (LSTM) neural network performs compared to a state of the art feature extraction neural network (FE-MLP) approach for the task of classifying PRI modulation. The results indicate that the proposed LSTM model performs consistently better than the FE-MLP approach across all tested noise levels. The downside of the proposed LSTM model is that it is significantly more complex than the FE-MLP approach. Future work could investigate if the LSTM model is too complex to use in a real world setting where computing power may be limited. Additionally, the LSTM model can, in a trivial manner, be modified to support more modulations than those tested in this work. Hence, future work could also evaluate how the proposed LSTM model performs when support for more modulations is added.<br>Radarsignaler används för att uppskatta plats, hastighet och riktning av objekt. Vissa radarer sänder ut signaler i form av pulser, medan andra sänder ut en kontinuerlig våg. Båda typer av radarer avger signaler enligt ett visst mönster, till exempel avger en pulsradar pulser med ett specifikt tidsintervall mellan pulserna. Detta tidsintervall kan antingen vara konstant, förändras linjärt, eller följa ett annat mönster. Intervallet mellan två pulser benämns ofta pulsrepetitionsintervall (PRI), och mönstret som definierar PRIn benämns ofta modulering. Att klassificera vilken PRI-modulering som används i en radarsignal är en viktig del i processen att identifiera vem som skickade ut signalen. Felaktig klassificering av den använda moduleringen kan leda till en felaktig gissning av identiteten av agenten som skickade ut signalen, vilket kan leda till ett dödligt utfall. Detta arbete undersöker hur väl det framtagna neurala nätverket som består av ett långt korttidsminne (LSTM) kan klassificera PRI-modulering i förhållande till en modern modell som använder särskilt utvalda beräknade särdrag från data och klassificerar dessa särdrag med ett neuralt nätverk. Resultaten indikerar att LSTM-modellen konsekvent klassificerar med högre träffsäkerhet än modellen som använder särdrag, vilket gäller för alla testade brusnivåer. Nackdelen med LSTM-modellen är att den är mer komplex än modellen som använder särdrag. Framtida arbete kan undersöka om LSTM-modellen är för komplex för att använda i ett verkligt scenario där beräkningskraften kan vara begränsad. Dessutom skulle framtida arbete kunna utvärdera hur väl LSTM-modellen kan klassificera PRI-moduleringar när stöd för fler moduleringar än de som testats i detta arbete läggs till, detta då stöd för ytterligare PRI-moduleringar kan läggas till i LSTM-modellen på ett trivialt sätt.
50
Díaz, González Fernando. "Federated Learning for Time Series Forecasting Using LSTM Networks: Exploiting Similarities Through Clustering." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-254665.
Full textAbstract:
Federated learning poses a statistical challenge when training on highly heterogeneous sequence data. For example, time-series telecom data collected over long intervals regularly shows mixed fluctuations and patterns. These distinct distributions are an inconvenience when a node not only plans to contribute to the creation of the global model but also plans to apply it on its local dataset. In this scenario, adopting a one-fits-all approach might be inadequate, even when using state-of-the-art machine learning techniques for time series forecasting, such as Long Short-Term Memory (LSTM) networks, which have proven to be able to capture many idiosyncrasies and generalise to new patterns. In this work, we show that by clustering the clients using these patterns and selectively aggregating their updates in different global models can improve local performance with minimal overhead, as we demonstrate through experiments using realworld time series datasets and a basic LSTM model.<br>Federated Learning utgör en statistisk utmaning vid träning med starkt heterogen sekvensdata. Till exempel så uppvisar tidsseriedata inom telekomdomänen blandade variationer och mönster över längre tidsintervall. Dessa distinkta fördelningar utgör en utmaning när en nod inte bara ska bidra till skapandet av en global modell utan även ämnar applicera denna modell på sin lokala datamängd. Att i detta scenario införa en global modell som ska passa alla kan visa sig vara otillräckligt, även om vi använder oss av de mest framgångsrika modellerna inom maskininlärning för tidsserieprognoser, Long Short-Term Memory (LSTM) nätverk, vilka visat sig kunna fånga komplexa mönster och generalisera väl till nya mönster. I detta arbete visar vi att genom att klustra klienterna med hjälp av dessa mönster och selektivt aggregera deras uppdateringar i olika globala modeller kan vi uppnå förbättringar av den lokal prestandan med minimala kostnader, vilket vi demonstrerar genom experiment med riktigt tidsseriedata och en grundläggande LSTM-modell.