To see the other types of publications on this topic, follow the link: Mathematical Models of Cognitive Processes and Neural Networks.
Journal articles on the topic 'Mathematical Models of Cognitive Processes and Neural Networks'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Mathematical Models of Cognitive Processes and Neural Networks.'
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 journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Marshakov, D. V. "Modeling of the processes of extracting rules from Neural Network components with Petri nets." Journal of Physics: Conference Series 2131, no. 2 (December 1, 2021): 022133. http://dx.doi.org/10.1088/1742-6596/2131/2/022133.
Full textAbstract:
Abstract The paper deals with the use of extended Petri nets in modeling the processes of extracting rules from neural network components. The mathematical model for extracting rules from neural network components based on a modified timed Petri net is constructed, followed by an analysis of its dynamic behavior based on a timed reachability graph, which is a set of all its states that can be reached when a finite number of transitions are fired. The proposed model allows us to move from the initial detailed structure to its simplified description, which preserves the possibility of obtaining information about the structure and dynamic behavior of the neural network system. The proposed approach can be used in the synthesis of cognitive systems with a neural network organization to provide computational support for the functions of forming, learning, and correcting cognitive networks that display neural network models.
2
Isomura, Takuya, and Karl Friston. "Reverse-Engineering Neural Networks to Characterize Their Cost Functions." Neural Computation 32, no. 11 (November 2020): 2085–121. http://dx.doi.org/10.1162/neco_a_01315.
Full textAbstract:
This letter considers a class of biologically plausible cost functions for neural networks, where the same cost function is minimized by both neural activity and plasticity. We show that such cost functions can be cast as a variational bound on model evidence under an implicit generative model. Using generative models based on partially observed Markov decision processes (POMDP), we show that neural activity and plasticity perform Bayesian inference and learning, respectively, by maximizing model evidence. Using mathematical and numerical analyses, we establish the formal equivalence between neural network cost functions and variational free energy under some prior beliefs about latent states that generate inputs. These prior beliefs are determined by particular constants (e.g., thresholds) that define the cost function. This means that the Bayes optimal encoding of latent or hidden states is achieved when the network's implicit priors match the process that generates its inputs. This equivalence is potentially important because it suggests that any hyperparameter of a neural network can itself be optimized—by minimization with respect to variational free energy. Furthermore, it enables one to characterize a neural network formally, in terms of its prior beliefs.
3
Musso, Mariel, Eva Kyndt, Eduardo Cascallar, and Filip Dochy. "Predicting Mathematical Performance: The Effect of Cognitive Processes and Self-Regulation Factors." Education Research International 2012 (2012): 1–13. http://dx.doi.org/10.1155/2012/250719.
Full textAbstract:
A substantial number of research studies have investigated the separate influence of working memory, attention, motivation, and learning strategies on mathematical performance and self-regulation in general. There is still little understanding of their impact on performance when taken together, understanding their interactions, and how much each of them contributes to the prediction of mathematical performance. With the emergence of new methodologies and technologies, such as the modelling with predictive systems, it is now possible to study these effects with approaches which use a wide range of data, including student characteristics, to estimate future performance without the need of traditional testing (Boekaerts and Cascallar, 2006). This research examines the different cognitive patterns and complex relations between cognitive variables, motivation, and background variables associated with different levels of mathematical performance using artificial neural networks (ANNs). A sample of 800 entering university students was used to develop three ANN models to identify the expected future level of performance in a mathematics test. These ANN models achieved high degree of precision in the correct classification of future levels of performance, showing differences in the pattern of relative predictive weight amongst those variables. The impact on educational quality, improvement, and accountability is highlighted.
4
Schubert, Anna-Lena, and Gidon T. Frischkorn. "Neurocognitive Psychometrics of Intelligence: How Measurement Advancements Unveiled the Role of Mental Speed in Intelligence Differences." Current Directions in Psychological Science 29, no. 2 (February 13, 2020): 140–46. http://dx.doi.org/10.1177/0963721419896365.
Full textAbstract:
More intelligent individuals typically show faster reaction times. However, individual differences in reaction times do not represent individual differences in a single cognitive process but in multiple cognitive processes. Thus, it is unclear whether the association between mental speed and intelligence reflects advantages in a specific cognitive process or in general processing speed. In this article, we present a neurocognitive-psychometrics account of mental speed that decomposes the relationship between mental speed and intelligence. We summarize research employing mathematical models of cognition and chronometric analyses of neural processing to identify distinct stages of information processing strongly related to intelligence differences. Evidence from both approaches suggests that the speed of higher-order processing is greater in smarter individuals, which may reflect advantages in the structural and functional organization of brain networks. Adopting a similar neurocognitive-psychometrics approach for other cognitive processes associated with intelligence (e.g., working memory or executive control) may refine our understanding of the basic cognitive processes of intelligence.
5
Wang, Yingxu. "The Cognitive Mechanisms and Formal Models of Consciousness." International Journal of Cognitive Informatics and Natural Intelligence 6, no. 2 (April 2012): 23–40. http://dx.doi.org/10.4018/jcini.2012040102.
Full textAbstract:
Consciousness is the sense of self and the sign of life in natural intelligence. One of the profound myths in cognitive informatics, psychology, brain science, and computational intelligence is how consciousness is generated by physiological organs and neural networks in the bran. This paper presents a formal model and a cognitive process of consciousness in order to explain how abstract consciousness is generated and what its cognitive mechanisms are. The hierarchical levels of consciousness are explored from the facets of neurology, physiology, and computational intelligence. A rigorous mathematical model of consciousness is created that elaborates the nature of consciousness. The cognitive process of consciousness is formally described using denotational mathematics. It is recognized that consciousness is a set of real-time mental information about bodily and emotional status of an individual stored in the cerebellums known as the Conscious Status Memory (CSM) and is processed/interpreted by the thalamus. The abstract intelligence model of consciousness can be applied in cognitive informatics, cognitive computing, and computational intelligence toward the mimicry and simulation of human perception and awareness of the internal states, external environment, and their interactions in reflexive, perceptive, cognitive, and instructive intelligence.
6
Gudowska-Nowak, E., M. A. Nowak, D. R. Chialvo, J. K. Ochab, and W. Tarnowski. "From Synaptic Interactions to Collective Dynamics in Random Neuronal Networks Models: Critical Role of Eigenvectors and Transient Behavior." Neural Computation 32, no. 2 (February 2020): 395–423. http://dx.doi.org/10.1162/neco_a_01253.
Full textAbstract:
The study of neuronal interactions is at the center of several big collaborative neuroscience projects (including the Human Connectome Project, the Blue Brain Project, and the Brainome) that attempt to obtain a detailed map of the entire brain. Under certain constraints, mathematical theory can advance predictions of the expected neural dynamics based solely on the statistical properties of the synaptic interaction matrix. This work explores the application of free random variables to the study of large synaptic interaction matrices. Besides recovering in a straightforward way known results on eigenspectra in types of models of neural networks proposed by Rajan and Abbott ( 2006 ), we extend them to heavy-tailed distributions of interactions. More important, we analytically derive the behavior of eigenvector overlaps, which determine the stability of the spectra. We observe that on imposing the neuronal excitation/inhibition balance, despite the eigenvalues remaining unchanged, their stability dramatically decreases due to the strong nonorthogonality of associated eigenvectors. This leads us to the conclusion that understanding the temporal evolution of asymmetric neural networks requires considering the entangled dynamics of both eigenvectors and eigenvalues, which might bear consequences for learning and memory processes in these models. Considering the success of free random variables theory in a wide variety of disciplines, we hope that the results presented here foster the additional application of these ideas in the area of brain sciences.
7
Crumpei, Gabriel, and Alina Gavriluţ. "Emergence, a Universal Phenomenon which Connects Reality to Consciousness, Natural Sciences to Humanities." Human and Social Studies 7, no. 2 (June 1, 2018): 89–106. http://dx.doi.org/10.2478/hssr-2018-0017.
Full textAbstract:
Abstract Progress in neuroscience has left a central question of psychism unanswered: what is consciousness? Modeling the psyche from a computational perspective has helped to develop cognitive neurosciences, but it has also shown their limits, of which the definition, description and functioning of consciousness remain essential. From Rene Descartes, who tackled the issue of psychism as the brain-mind dualism, to Chambers, who defined qualia as the tough, difficult problem of research in neuroscience, many hypotheses and theories have been issued to encompass the phenomenon of consciousness. Neuroscience specialists, such as Giulio Tononi or David Eagleman, consider consciousness as a phenomenon of emergence of all processes that take place in the brain. This hypothesis has the advantage of being supported by progress made in the study of complex systems in which the issue of emergence can be mathematically formalized and analyzed by physical-mathematical models. The current tendency to associate neural networks within the broad scope of network science also allows for a physical-mathematical formalization of phenomenology in neural networks and the construction of information-symbolic models. The extrapolation of emergence at the level of physical systems, biological systems and psychic systems can bring new models that can also be applied to the concept of consciousness. The meaning and significance that seem to structure the nature of consciousness is found as direction of evolution and teleological finality, of integration in the whole system and in any complex system at all scales. Starting from the wave-corpuscle duality in quantum physics, we can propose a model for structuring reality, based on the emergence of systems that contribute to the integration and coherence of the entire reality. Physical-mathematical models based mainly on (mereo)topology can provide a mathematical formalization path, and the paradigm of information could allow the development of a pattern of emergence, that is common to all systems, including the psychic system, the difference being given only by the degree of information complexity. Thus, the mind-brain duality, which has been dominating the representation on psychism for a few centuries, could be solved by an informational approach, describing the connection between object and subject, reality and human consciousness, between mind and brain, thus unifying the perspective on natural sciences and humanities.
8
HOTTON, SCOTT, and JEFF YOSHIMI. "THE DYNAMICS OF EMBODIED COGNITION." International Journal of Bifurcation and Chaos 20, no. 04 (April 2010): 943–72. http://dx.doi.org/10.1142/s0218127410026241.
Full textAbstract:
Historically cognition was understood as the result of processes occurring solely in the brain. Recently, however, cognitive scientists and philosophers studying "embodied" or "situated" cognition have begun emphasizing the role of the body and environment in which brains are situated, i.e. they view the brain as an "open system". However, these theorists frequently rely on dynamical systems which are traditionally viewed as closed systems. We address this tension by extending the framework of dynamical systems theory. We show how structures which appear in the state space of an embodied agent differ from those that appear in closed systems, and we show how these structures can be used to model representational processes in embodied agents. We focus on neural networks as models of embodied cognition.
9
Zhuravskyi, Yurii, Oleg Sova, Serhii Korobchenko, Vitaliy Baginsky, Yurii Tsimura, Leonid Kolodiichuk, Pavlo Khomenko, Nataliia Garashchuk, Olena Orobinska, and Andrii Shyshatskyi. "Development of object state evaluation method in intelligent decision support systems." Eastern-European Journal of Enterprise Technologies 6, no. 9 (114) (December 29, 2021): 54–63. http://dx.doi.org/10.15587/1729-4061.2021.246421.
Full textAbstract:
Accurate and objective object analysis requires multi-parameter estimation with significant computational costs. A methodological approach to improve the accuracy of assessing the state of the monitored object is proposed. This methodological approach is based on a combination of fuzzy cognitive models, advanced genetic algorithm and evolving artificial neural networks. The methodological approach has the following sequence of actions: building a fuzzy cognitive model; correcting the fuzzy cognitive model and training knowledge bases. The distinctive features of the methodological approach are that the type of data uncertainty and noise is taken into account while constructing the state of the monitored object using fuzzy cognitive models. The novelties while correcting fuzzy cognitive models using a genetic algorithm are taking into account the type of data uncertainty, taking into account the adaptability of individuals to iteration, duration of the existence of individuals and topology of the fuzzy cognitive model. The advanced genetic algorithm increases the efficiency of correcting factors and the relationships between them in the fuzzy cognitive model. This is achieved by finding solutions in different directions by several individuals in the population. The training procedure consists in learning the synaptic weights of the artificial neural network, the type and parameters of the membership function and the architecture of individual elements and the architecture of the artificial neural network as a whole. The use of the method allows increasing the efficiency of data processing at the level of 16–24 % using additional advanced procedures. The proposed methodological approach should be used to solve the problems of assessing complex and dynamic processes characterized by a high degree of complexity.
10
Pavlic, Tomislav, Krunoslav Kušec, Danijel Radočaj, Alen Britvić, Marin Lukas, Vladimir Milić, and Mladen Crneković. "Cognitive Model of the Closed Environment of a Mobile Robot Based on Measurements." Applied Sciences 11, no. 6 (March 20, 2021): 2786. http://dx.doi.org/10.3390/app11062786.
Full textAbstract:
In recent years in mobile robotics, the focus has been on methods, in which the fusion of measurement data from various systems leads to models of the environment that are of a probabilistic type. The cognitive model of the environment is less accurate than the exact mathematical one, but it is unavoidable in the robot collaborative interaction with a human. The subject of the research proposed in this paper is the development of a model for learning and planning robot operations. The task of operations and mapping the unknown environment, similar to how humans do the same tasks in the same conditions has been explored. The learning process is based on a virtual dynamic model of a mobile robot, identical to a real mobile robot. The mobile robot’s motion with developed artificial neural networks and genetic algorithms is defined. The transfer method of obtained knowledge from simulated to a real system (Sim-To-Real; STR) is proposed. This method includes a training step, a simultaneous reasoning step, and an application step of trained and learned knowledge to control a real robot’s motion. Use of the basic cognitive elements language, a robot’s environment, and its correlation to that environment is described. Based on that description, a higher level of information about the mobile robot’s environment is obtained. The information is directly generated by the fusion of measurement data obtained from various systems.
11
Zacarias-Morales, Noel, Pablo Pancardo, José Adán Hernández-Nolasco, and Matias Garcia-Constantino. "Attention-Inspired Artificial Neural Networks for Speech Processing: A Systematic Review." Symmetry 13, no. 2 (January 28, 2021): 214. http://dx.doi.org/10.3390/sym13020214.
Full textAbstract:
Artificial Neural Networks (ANNs) were created inspired by the neural networks in the human brain and have been widely applied in speech processing. The application areas of ANN include: Speech recognition, speech emotion recognition, language identification, speech enhancement, and speech separation, amongst others. Likewise, given that speech processing performed by humans involves complex cognitive processes known as auditory attention, there has been a growing amount of papers proposing ANNs supported by deep learning algorithms in conjunction with some mechanism to achieve symmetry with the human attention process. However, while these ANN approaches include attention, there is no categorization of attention integrated into the deep learning algorithms and their relation with human auditory attention. Therefore, we consider it necessary to have a review of the different ANN approaches inspired in attention to show both academic and industry experts the available models for a wide variety of applications. Based on the PRISMA methodology, we present a systematic review of the literature published since 2000, in which deep learning algorithms are applied to diverse problems related to speech processing. In this paper 133 research works are selected and the following aspects are described: (i) Most relevant features, (ii) ways in which attention has been implemented, (iii) their hypothetical relationship with human attention, and (iv) the evaluation metrics used. Additionally, the four publications most related with human attention were analyzed and their strengths and weaknesses were determined.
12
Nie, Yimin, and Masami Tatsuno. "Information-Geometric Measures for Estimation of Connection Weight Under Correlated Inputs." Neural Computation 24, no. 12 (December 2012): 3213–45. http://dx.doi.org/10.1162/neco_a_00367.
Full textAbstract:
The brain processes information in a highly parallel manner. Determination of the relationship between neural spikes and synaptic connections plays a key role in the analysis of electrophysiological data. Information geometry (IG) has been proposed as a powerful analysis tool for multiple spike data, providing useful insights into the statistical interactions within a population of neurons. Previous work has demonstrated that IG measures can be used to infer the connection weight between two neurons in a neural network. This property is useful in neuroscience because it provides a way to estimate learning-induced changes in synaptic strengths from extracellular neuronal recordings. A previous study has shown, however, that this property would hold only when inputs to neurons are not correlated. Since neurons in the brain often receive common inputs, this would hinder the application of the IG method to real data. We investigated the two-neuron-IG measures in higher-order log-linear models to overcome this limitation. First, we mathematically showed that the estimation of uniformly connected synaptic weight can be improved by taking into account higher-order log-linear models. Second, we numerically showed that the estimation can be improved for more general asymmetrically connected networks. Considering the estimated number of the synaptic connections in the brain, we showed that the two-neuron IG measure calculated by the fourth- or fifth-order log-linear model would provide an accurate estimation of connection strength within approximately a 10% error. These studies suggest that the two-neuron IG measure with higher-order log-linear expansion is a robust estimator of connection weight even under correlated inputs, providing a useful analytical tool for real multineuronal spike data.
13
Tóth, Brigitta, Ádám Boncz, Bálint File, István Winkler, and Márk Molnár. "A humán agyi aktivitás hálózatelemzési modellezése – Humán agyi hálózatok." Scientia et Securitas 1, no. 1 (December 17, 2020): 21–28. http://dx.doi.org/10.1556/112.2020.00004.
Full textAbstract:
Összefoglalás. A hálózatkutatás idegtudományi alkalmazása áttörő eredményt hozott a humán kogníció és a neurális rendszerek közötti kapcsolat megértésében. Jelen tanulmány célja a neurális hálózatok néhány kutatási területét mutatja be a laborunkban végzett vizsgálatok eredményein keresztül. Bemutatjuk az agyi aktivitás mérésének és az agyi területek közötti kommunikációs hálózatok modellezésének technikáját. Majd kiemelünk két kutatási terület: 1) az agyi hálózatok életkori változásainak vizsgálatát, ami választ ad arra, hogy hogyan öregszik az emberi agy; 2) az emberi agyak közötti hálózat modelljének vizsgálatát, amely a hatékony emberi kommunikáció idegrendszeri mechanizmusait próbálja feltárni. Tárgyaljuk a humán kommunikációra képes mesterséges intelligencia fejlesztésének lehetőségét is. Végül kitérünk az agyi hálózatok kutatásának biztonságpolitikai vonatkozásaira. Summary. The human brain consists of 100 billion neurons connected by about 100 trillion synapses, which are hierarchically organized in different scales in anatomical space and time. Thus, it sounds reasonable to assume that the brain is the most complex network known to man. Network science applications in neuroscience are aimed to understand how human feeling, thought and behavior could emerge from this biological system of the brain. The present review focuses on the recent results and the future of network neuroscience. The following topics will be discussed: Modeling the network of communication among brain areas. Neural activity can be recorded with high temporal precision using electroencephalography (EEG). Communication strength between brain regions then might be estimated by calculating mathematical synchronization indices between source localized EEG time series. Finally, graph theoretical models can describe the relationship between system elements (i.e. efficiency of communication or centrality of an element). How does the brain age? While for a newborn the high plasticity of the brain provides the foundation of cognitive development, cognition declines with advanced age due to so far largely unknown neural mechanisms. In one of our studies, we demonstrated that there is a correlation between the anatomical development of the brain (at prenatal age) and its network topology. Specifically, the more developed the baby’s brain, the more functionally specialized/modular it was. In another study we found that in older adults, when compared to young adults, connectivity within modules of their brain network is decreased, with an associated decline in their short-term memory capacity. Moreover, Mild Cognitive Impairment patients (early stage of Alzheimer) were characterized with a significantly lower level of connectivity between their brain modules than the healthy elderly. Human communication via shared network of brain activity. In another study we recorded the brain activity of a speaker and multiple listeners. We investigated the brain network similarity across listeners and between the speaker and listeners. We found that brain activity was significantly correlated among listeners, providing evidence for the fact that the same content is processed via similar neural computations within different brains. The data also suggested that the more the brain activity synchronizes the more the mental state of the individuals overlap. We also found significantly synchronized brain activity between speaker and listeners. Specifically 1) listeners’ brain activity within the speech processing cortices was synchronized to speaker’s brain activity with a time lag, indicating that listeners’ speech comprehension processes replicated the speaker’s speech production processes; and 2) listeners’ frontal cortical activity was synchronized to speaker’s later brain activity, that is, listeners preceded the speaker, indicating that speech content is predicted by the listeners based on the context. Future challenges. Future research could target artificial intelligence development that is capable of human-like communication. To achieve this, the simultaneous recording of brain activity from listener and speaker is needed together with efficiency of the communication. These data could be then modelled via AI to detect biomarkers of communication efficiency. In general, neurotechnology has been rapidly developing within and outside of research and in clinical fields thus it is time for re-conceptualizing the corresponding human right law in order to avoid unwanted consequences of technological applications.
14
Glukhikh, V. A., S. M. Eliseev, and N. P. Kirsanova. "Artificial Intelligence as a Problem of Modern Sociology." Discourse 8, no. 1 (February 25, 2022): 82–93. http://dx.doi.org/10.32603/2412-8562-2022-8-1-82-93.
Full textAbstract:
Introduction. There are currently many different methods in the field of artificial intelligence research. Methods of mathematical, cognitive and philosophical sciences are dominant among them. All research approaches are united by the hypothesis that natural and artificial intelligence are fundamentally comparable. The sociality formed as a result of these changes attracts increasing attention from both foreign and Russian researchers. The purpose of this article is to clarify the theoretical and methodological approaches in the study of artificial intelligence in the social sciences, especially in sociology.Methodology and sources. The article is based on an interdisciplinary approach, which allows outlining the scale of the research problem, coordinating the methodological approach to the organization of research, smoothing the contradictions of ideas and categories, which are operated by different sciences in the study of artificial intelligence.Results and discussion. According to the authors, the widely used concept of Artificial Intelligence is more a scientific metaphor than a proven empirical fact. Currently there is no such thing as artificial intelligence. There are neural networks, machine learning, which can solve certain problems in the real world. Artificial intelligence is a metaphor that captures a certain level of human knowledge about the introduction of information technology, based on computer hardware and specialized software. To treat artificial intelligence as an empirical fact is a fallacy that is not appropriate in science.Conclusion. Sociology is only taking its first steps in the field of artificial intelligence research. It does not have its own methodological tools for analyzing artificial intelligence and the social reality that arises from its introduction into the everyday life of society. Artificial Intelligence changes people's daily lives, embedding itself in everyday social practices, and forming a hybrid social world for the social sciences to study. Today there is a debate about the place of artificial intelligence in sociology. According to the authors, sociological fantasies and speculations are not appropriate here. In order to correctly and accurately define the problem of artificial intelligence in the social sciences, it is necessary to carefully analyze the opinions of the experts in the exact sciences, in which artificial intelligence is understood as algorithms or models created by human, and which perform certain tasks and help them manage specific processes in various spheres of society.
15
Elliott, T., C. I. Howarth, and N. R. Shadbolt. "Axonal Processes and Neural Plasticity: A Reply." Neural Computation 10, no. 3 (April 1, 1998): 549–54. http://dx.doi.org/10.1162/089976698300017656.
Full textAbstract:
We examine the claim that a class of sprouting-and-retraction models is mathematically equivalent to a fixed-anatomy model. We accept, subject to important caveats, a narrow mathematical equivalence of the energy functions in both classes of model. We argue that this narrow equivalence of energy functions does not, however, entail equivalence of the models. Indeed, the claim of complete model equivalence hides significant dynamical differences between the approaches, which we discuss. We also disagree that our work demonstrates that subtractive constraint enforcement is natural in fixed-anatomy models.
16
Harmati, István Á. "Dynamics of Fuzzy-Rough Cognitive Networks." Symmetry 13, no. 5 (May 15, 2021): 881. http://dx.doi.org/10.3390/sym13050881.
Full textAbstract:
Fuzzy-rough cognitive networks (FRCNs) are interpretable recurrent neural networks, primarily designed for solving classification problems. Their structure is simple and transparent, while the performance is comparable to the well-known black-box classifiers. Although there are many applications on fuzzy cognitive maps and recently for FRCNS, only a very limited number of studies discuss the theoretical issues of these models. In this paper, we examine the behaviour of FRCNs viewing them as discrete dynamical systems. It will be shown that their mathematical properties highly depend on the size of the network, i.e., there are structural differences between the long-term behaviour of FRCN models of different size, which may influence the performance of these modelling tools.
17
Gridin, V. N., V. V. Doenin, V. V. Panishchev, and I. S. Razzhivaykin. "Digital Model: Behavior Forecast in Transport Processes." World of Transport and Transportation 17, no. 2 (September 13, 2019): 6–14. http://dx.doi.org/10.30932/1992-3252-2019-17-2-6-14.
Full textAbstract:
In today’s world, many processes and events depend on forecasting. With development of mathematical models, an increasing number of factors influencing the final result of the forecast are taken into account, which in turn leads to the use of neural networks. But for training a neural network, source data sets are required, which are often not always sufficient or may not exist at all. The article describes a method of obtaining information as close to reality as possible. The proposed approach is to generate input data using simulation models of an object. The solution of a problem of generation of data sets and of training of a neural network is shown at the example of a typical marshalling railway station, and of a simulation of operations of a shunting hump. The considered examples confirmed the validity of the proposed methodological approach to generation of source data for neural networks using simulation models of a real object, based on a digital mathematical model, which makes it possible to obtain a simulation model of movement of transport objects, which is reliable in forecasting transport processes and creating relevant control algorithms.
18
Lewis, John E., and Leon Glass. "Nonlinear Dynamics and Symbolic Dynamics of Neural Networks." Neural Computation 4, no. 5 (September 1992): 621–42. http://dx.doi.org/10.1162/neco.1992.4.5.621.
Full textAbstract:
A piecewise linear equation is proposed as a method of analysis of mathematical models of neural networks. A symbolic representation of the dynamics in this equation is given as a directed graph on an N-dimensional hypercube. This provides a formal link with discrete neural networks such as the original Hopfield models. Analytic criteria are given to establish steady states and limit cycle oscillations independent of network dimension. Model networks that display multiple stable limit cycles and chaotic dynamics are discussed. The results show that such equations are a useful and efficient method of investigating the behavior of neural networks.
19
Miller, Kenneth D., and Francesco Fumarola. "Mathematical Equivalence of Two Common Forms of Firing Rate Models of Neural Networks." Neural Computation 24, no. 1 (January 2012): 25–31. http://dx.doi.org/10.1162/neco_a_00221.
Full textAbstract:
We demonstrate the mathematical equivalence of two commonly used forms of firing rate model equations for neural networks. In addition, we show that what is commonly interpreted as the firing rate in one form of model may be better interpreted as a low-pass-filtered firing rate, and we point out a conductance-based firing rate model.
20
Scott, Gary M., and W. Harmon Ray. "Neural Network Process Models Based on Linear Model Structures." Neural Computation 6, no. 4 (July 1994): 718–38. http://dx.doi.org/10.1162/neco.1994.6.4.718.
Full textAbstract:
The KBANN (Knowledge-Based Artificial Neural Networks) approach uses neural networks to refine knowledge that can be written in the form of simple propositional rules. This idea is extended by presenting the MANNIDENT (Multivariable Artificial Neural Network Identification) algorithm by which the mathematical equations of linear dynamic process models determine the topology and initial weights of a network, which is further trained using backpropagation. This method is applied to the task of modeling a nonisothermal chemical reactor in which a first-order exothermic reaction is occurring. This method produces statistically significant gains in accuracy over both a standard neural network approach and a linear model. Furthermore, using the approximate linear model to initialize the weights of the network produces statistically less variation in model fidelity. By structuring the neural network according to the approximate linear model, the model can be readily interpreted.
21
Herd, Seth A., Kai A. Krueger, Trenton E. Kriete, Tsung-Ren Huang, Thomas E. Hazy, and Randall C. O'Reilly. "Strategic Cognitive Sequencing: A Computational Cognitive Neuroscience Approach." Computational Intelligence and Neuroscience 2013 (2013): 1–18. http://dx.doi.org/10.1155/2013/149329.
Full textAbstract:
We address strategic cognitive sequencing, the “outer loop” of human cognition: how the brain decides what cognitive process to apply at a given moment to solve complex, multistep cognitive tasks. We argue that this topic has been neglected relative to its importance for systematic reasons but that recent work on how individual brain systems accomplish their computations has set the stage for productively addressing how brain regions coordinate over time to accomplish our most impressive thinking. We present four preliminary neural network models. The first addresses how the prefrontal cortex (PFC) and basal ganglia (BG) cooperate to perform trial-and-error learning of short sequences; the next, how several areas of PFC learn to make predictions of likely reward, and how this contributes to the BG making decisions at the level of strategies. The third models address how PFC, BG, parietal cortex, and hippocampus can work together to memorize sequences of cognitive actions from instruction (or “self-instruction”). The last shows how a constraint satisfaction process can find useful plans. The PFC maintains current and goal states and associates from both of these to find a “bridging” state, an abstract plan. We discuss how these processes could work together to produce strategic cognitive sequencing and discuss future directions in this area.
22
Galerkin, Y. B., A. G. Nikiforov, O. A. Solovyeva, and E. Y. Popova. "Simulating Characteristics of Vaneless Diffusers Using Neural Networks." Proceedings of Higher Educational Institutions. Маchine Building, no. 07 (724) (July 2020): 29–42. http://dx.doi.org/10.18698/0536-1044-2020-7-29-42.
Full textAbstract:
To calculate flow parameters of a vaneless diffuser of the centrifugal compressor stage, it is sufficient to determine the loss coefficient and the flow direction at the outlet. The paper presents the results of modeling the characteristics of these two parameters using neural networks and CFD methods. To obtain mathematical models, ANSYS calculation data was used for vaneless diffusers with a relative width of 0.014–0.1, relative outlet diameter of 1.4–2.0, inlet flow angle of 10–90° and velocity coefficient of 0.39–0.82, with the Reynolds number being in the range of 87 500–1 030 000. A comparison with the theory showed the regularity of gas-dynamic characteristics, and comparison with well-known experiments showed the correspondence of the flow structure. In order to improve the accuracy of simulation using neural networks, various recommendations on the preparation and processing of the initial data were collected and tested: identification of conflict examples and outliers, data normalization, improving the quality of the neural network training under the insufficient amount of sampling, etc. The application of the aforementioned recommendations significantly improved the accuracy of simulation. A simulation experiment based on neural models for studying the influence of dimensions, diffuser shape and similarity criteria on the diffuser gas dynamic characteristics made it possible to verify physical adequacy of the mathematical models, obtain new data on energy conversion processes and produce a set of recommendations on the optimal design of vaneless diffusers.
23
García-Planas, Maria Isabel, and Maria Victoria García-Camba. "Controllability of Brain Neural Networks in Learning Disorders—A Geometric Approach." Mathematics 10, no. 3 (January 21, 2022): 331. http://dx.doi.org/10.3390/math10030331.
Full textAbstract:
The human brain can be interpreted mathematically as a linear dynamical system that shifts through various cognitive regions promoting more or less complicated behaviors. The dynamics of brain neural network play a considerable role in cognitive function and therefore of interest in the bid to understand the learning processes and the evolution of possible disorders. The mathematical theory of systems and control makes available procedures, concepts, and criteria that can be applied to ease the perception of the dynamic processes that administer the evolution of the brain with learning and its control with treatment in case of disorder. In this work, a geometric study through the conception of exact controllability is comprehended to detect the minimum set and the location of the driving nodes of learning. We will describe the different roles of the nodes in the control of the paths of brain networks and show the transition of some driving nodes and the preservation of the rest in the course of learning in patients with some learning disability.
24
Lo, James Ting-Ho. "A Low-Order Model of Biological Neural Networks." Neural Computation 23, no. 10 (October 2011): 2626–82. http://dx.doi.org/10.1162/neco_a_00166.
Full textAbstract:
A biologically plausible low-order model (LOM) of biological neural networks is proposed. LOM is a recurrent hierarchical network of models of dendritic nodes and trees; spiking and nonspiking neurons; unsupervised, supervised covariance and accumulative learning mechanisms; feedback connections; and a scheme for maximal generalization. These component models are motivated and necessitated by making LOM learn and retrieve easily without differentiation, optimization, or iteration, and cluster, detect, and recognize multiple and hierarchical corrupted, distorted, and occluded temporal and spatial patterns. Four models of dendritic nodes are given that are all described as a hyperbolic polynomial that acts like an exclusive-OR logic gate when the model dendritic nodes input two binary digits. A model dendritic encoder that is a network of model dendritic nodes encodes its inputs such that the resultant codes have an orthogonality property. Such codes are stored in synapses by unsupervised covariance learning, supervised covariance learning, or unsupervised accumulative learning, depending on the type of postsynaptic neuron. A masking matrix for a dendritic tree, whose upper part comprises model dendritic encoders, enables maximal generalization on corrupted, distorted, and occluded data. It is a mathematical organization and idealization of dendritic trees with overlapped and nested input vectors. A model nonspiking neuron transmits inhibitory graded signals to modulate its neighboring model spiking neurons. Model spiking neurons evaluate the subjective probability distribution (SPD) of the labels of the inputs to model dendritic encoders and generate spike trains with such SPDs as firing rates. Feedback connections from the same or higher layers with different numbers of unit-delay devices reflect different signal traveling times, enabling LOM to fully utilize temporally and spatially associated information. Biological plausibility of the component models is discussed. Numerical examples are given to demonstrate how LOM operates in retrieving, generalizing, and unsupervised and supervised learning.
25
Dizaji, S. Haleh S., Saeid Pashazadeh, and Javad Musevi Niya. "A Comparative Study of Some Point Process Models for Dynamic Networks." Complexity 2022 (September 16, 2022): 1–21. http://dx.doi.org/10.1155/2022/1616116.
Full textAbstract:
Modeling dynamic networks has attracted much interest in recent years, which helps understand networks’ behavior. Many works have been dedicated to modeling discrete-time networks, but less work is done for continuous-time networks. Point processes as powerful tools for modeling discrete events in continuous time have been widely used for modeling events over networks and their dynamics. These models have solid mathematical assumptions, making them interpretable but decreasing their generalizability for different datasets. Hence, neural point processes were introduced that don’t have strong assumptions on generative functions. However, these models can be impractical in the case of a large number of event types. This research presents a comparative study of different point process (Hawkes) models for continuous-time networks. Furthermore, a previously introduced neural point process (neural Hawkes) model is applied for modeling network interactions. In this work, network clustering is used for specifying interaction types. These methods are compared using different synthetic and real-world datasets, and their efficiency is evaluated on these datasets. The experiments represent that each model is appropriate for a group of datasets. In addition, the effect of clustering on results is discussed, and experiments for different clusters are presented.
26
Wang, Yingxu, Lotfi A. Zadeh, Bernard Widrow, Newton Howard, Françoise Beaufays, George Baciu, D. Frank Hsu, et al. "Abstract Intelligence." International Journal of Cognitive Informatics and Natural Intelligence 11, no. 1 (January 2017): 1–15. http://dx.doi.org/10.4018/ijcini.2017010101.
Full textAbstract:
Basic studies in denotational mathematics and mathematical engineering have led to the theory of abstract intelligence (aI), which is a set of mathematical models of natural and computational intelligence in cognitive informatics (CI) and cognitive computing (CC). Abstract intelligence triggers the recent breakthroughs in cognitive systems such as cognitive computers, cognitive robots, cognitive neural networks, and cognitive learning. This paper reports a set of position statements presented in the plenary panel (Part II) of IEEE ICCI*CC'16 on Cognitive Informatics and Cognitive Computing at Stanford University. The summary is contributed by invited panelists who are part of the world's renowned scholars in the transdisciplinary field of CI and CC.
27
Abitova, G., V. Nikulin, and T. Zadenova. "NEURAL NETWORK MODELING AND OPTIMISING OF THE AGGLOMERATION PROCESS OF SULPHIDE POLYMETALLIC ORES." Scientific Journal of Astana IT University, no. 6 (June 30, 2021): 4–14. http://dx.doi.org/10.37943/aitu.2021.76.49.001.
Full textAbstract:
During the operation of the lead-zinc production while processing of polymetallic ores, problems arose related to the quality of products and the efficient use of equipment – agglomeration furnace and crushing apparatus. Previously, such issues were resolved due to the experiences and based on mathematical modeling of processes. The mathematical model for optimizing unnecessary such operating mode is a difficult program. Performing calculations is required a fairly large investment of time and resources. Therefore, the program of the mathematical model for optimizing the operating mode of the agglomeration furnace and the crushing device for sinter firing was replaced with a neural network by implementing the process of training the network based on the results of calculations on a mathematical model. The results obtained showed that neural network models were more accurate than mathematical models, which made it possible to solve production optimization problems of great complexity. The use of neural networks for modeling technological processes has made it possible to increase the efficiency of product quality control systems and automatic control systems for the roasting of sulfide polymetallic ores.
28
Mikkilineni, Rao. "Information Processing, Information Networking, Cognitive Apparatuses and Sentient Software Systems." Proceedings 47, no. 1 (May 13, 2020): 27. http://dx.doi.org/10.3390/proceedings2020047027.
Full textAbstract:
In the physical world, computing processes, message communication networks and cognitive apparatuses are the building blocks of sentient beings. Genes and neural networks provide complementary information processing models that enable execution of mechanisms dealing with “life” using physical, chemical and biological processes. A cognizing agent architecture (mind) provides the orchestration of body and the brain to manage the “life” processes to deal with fluctuations and maintain survival and sustenance. We present a new information processing architecture that enables “digital genes” and “digital neurons” with cognizing agent architecture to design and implement sentient, resilient and intelligent systems in the digital world.
29
Mikkilineni, Rao. "Information Processing, Information Networking, Cognitive Apparatuses and Sentient Software Systems." Proceedings 47, no. 1 (May 13, 2020): 27. http://dx.doi.org/10.3390/proceedings47010027.
Full textAbstract:
In the physical world, computing processes, message communication networks and cognitive apparatuses are the building blocks of sentient beings. Genes and neural networks provide complementary information processing models that enable execution of mechanisms dealing with “life” using physical, chemical and biological processes. A cognizing agent architecture (mind) provides the orchestration of body and the brain to manage the “life” processes to deal with fluctuations and maintain survival and sustenance. We present a new information processing architecture that enables “digital genes” and “digital neurons” with cognizing agent architecture to design and implement sentient, resilient and intelligent systems in the digital world.
30
Schmitt, Oliver, Christian Nitzsche, Peter Eipert, Vishnu Prathapan, Marc-Thorsten Hütt, and Claus Hilgetag. "Reaction-diffusion models in weighted and directed connectomes." PLOS Computational Biology 18, no. 10 (October 28, 2022): e1010507. http://dx.doi.org/10.1371/journal.pcbi.1010507.
Full textAbstract:
Connectomes represent comprehensive descriptions of neural connections in a nervous system to better understand and model central brain function and peripheral processing of afferent and efferent neural signals. Connectomes can be considered as a distinctive and necessary structural component alongside glial, vascular, neurochemical, and metabolic networks of the nervous systems of higher organisms that are required for the control of body functions and interaction with the environment. They are carriers of functional epiphenomena such as planning behavior and cognition, which are based on the processing of highly dynamic neural signaling patterns. In this study, we examine more detailed connectomes with edge weighting and orientation properties, in which reciprocal neuronal connections are also considered. Diffusion processes are a further necessary condition for generating dynamic bioelectric patterns in connectomes. Based on our high-precision connectome data, we investigate different diffusion-reaction models to study the propagation of dynamic concentration patterns in control and lesioned connectomes. Therefore, differential equations for modeling diffusion were combined with well-known reaction terms to allow the use of connection weights, connectivity orientation and spatial distances. Three reaction-diffusion systems Gray-Scott, Gierer-Meinhardt and Mimura-Murray were investigated. For this purpose, implicit solvers were implemented in a numerically stable reaction-diffusion system within the framework of neuroVIISAS. The implemented reaction-diffusion systems were applied to a subconnectome which shapes the mechanosensitive pathway that is strongly affected in the multiple sclerosis demyelination disease. It was found that demyelination modeling by connectivity weight modulation changes the oscillations of the target region, i.e. the primary somatosensory cortex, of the mechanosensitive pathway. In conclusion, a new application of reaction-diffusion systems to weighted and directed connectomes has been realized. Because the implementation were performed in the neuroVIISAS framework many possibilities for the study of dynamic reaction-diffusion processes in empirical connectomes as well as specific randomized network models are available now.
31
Shevchenko, Alla, Andrey Shevchenko, O. Tyatyushkina, and S. Ulyanov. "Intelligent robust controller based on cognitive computing technologies. Pt. 1: cognitive Control models with THE BRAIN emotional learning." System Analysis in Science and Education, no. 4 (2020) (December 30, 2020): 90–134. http://dx.doi.org/10.37005/2071-9612-2020-4-90-134.
Full textAbstract:
n on-linecontrol and decision-making systems, emotional brain training is a preferred methodology (compared to stochastic gradient-based and evolutionary algorithms) due to its low computational complexity and fast robust learning. To describe the emotional learning of the brain, a mathematical model was created —the brain emotional learning controller (BELC). The design of intelligent systems based on emotional signals basedoncontrol methods assoft computing technologies: artificial neural networks, fuzzy control and genetic algorithms. Based on the simulated mathematical model of mammals BEL, a controller architecture has been developed. Applied approachcalled “Brain Emotional Learning Based Intelligent Controller” (BELBIC) —a neurobiologically motivated intelligent controller based on a computational model of emotional learning in the mammalian limbic system. The article describes applied models of intelligent regulators based on emotional learning of the brain. BELBIC's learning capabilities;versatility and low computational complexity make it a very promising toolkitfor on-lineapplications.
32
Bathellier, Brice, Alan Carleton, and Wulfram Gerstner. "Gamma Oscillations in a Nonlinear Regime: A Minimal Model Approach Using Heterogeneous Integrate-and-Fire Networks." Neural Computation 20, no. 12 (December 2008): 2973–3002. http://dx.doi.org/10.1162/neco.2008.11-07-636.
Full textAbstract:
Fast oscillations and in particular gamma-band oscillation (20–80 Hz) are commonly observed during brain function and are at the center of several neural processing theories. In many cases, mathematical analysis of fast oscillations in neural networks has been focused on the transition between irregular and oscillatory firing viewed as an instability of the asynchronous activity. But in fact, brain slice experiments as well as detailed simulations of biological neural networks have produced a large corpus of results concerning the properties of fully developed oscillations that are far from this transition point. We propose here a mathematical approach to deal with nonlinear oscillations in a network of heterogeneous or noisy integrate-and-fire neurons connected by strong inhibition. This approach involves limited mathematical complexity and gives a good sense of the oscillation mechanism, making it an interesting tool to understand fast rhythmic activity in simulated or biological neural networks. A surprising result of our approach is that under some conditions, a change of the strength of inhibition only weakly influences the period of the oscillation. This is in contrast to standard theoretical and experimental models of interneuron network gamma oscillations (ING), where frequency tightly depends on inhibition strength, but it is similar to observations made in some in vitro preparations in the hippocampus and the olfactory bulb and in some detailed network models. This result is explained by the phenomenon of suppression that is known to occur in strongly coupled oscillating inhibitory networks but had not yet been related to the behavior of oscillation frequency.
33
Butusov, O. B., O. P. Nikiforova, and N. I. Redikultseva. "Mathematical methods for the analysis of migration processes on the basis of demographic data." Izvestiya MGTU MAMI 9, no. 1-4 (July 10, 2015): 21–25. http://dx.doi.org/10.17816/2074-0530-67102.
Full textAbstract:
The problem of mathematical analysis of demographic data was investigated for data, where information on migration flows is taken into account implicitly. Regression techniques, neural networks and stochastic analysis were used for the mathematical analysis of demographic processes. Two age groups were considered: young (0 - 39 years) and elderly (40 - 70 years). While development of stochastic models the theory of Markov chains and transition matrix were used. The parameterization and model identification were conducted according to Rosstat data.
34
Nikiforov, Aleksandr, Alexei Kuchumov, Sergei Terentev, Evgeniy Petukhov, and Kirill Kabalyk. "Simulation of gas-dynamic characteristics of a centrifugal compressor vane diffuser using neural networks." E3S Web of Conferences 140 (2019): 05003. http://dx.doi.org/10.1051/e3sconf/201914005003.
Full textAbstract:
The paper presents the results of mathematical simulation of the characteristics of a vane diffuser of a centrifugal compressor intermediate stage, such as the loss coefficient and the deviation angle versus the outlet vane angle of the diffuser. The simulation of these characteristics was made on the basis of processing the results of studies performed by the Research Laboratory “Gas Dynamics of Turbomachines” of Peter the Great St.Petersburg Polytechnic University at the model characteristics of vane diffusers. Given the almost complete absence of recommendations in the literature, the paper describes the technology for constructing neural network models, which includes preparing a sample of input data and determining the optimal structure of the neural network. Based on the obtained mathematical models, a computational experiment was carried out in order to determine the influence of the main geometric and gas-dynamic parameters on the efficiency of vane diffusers. The results of the computational experiment on neural models of the efficiency of a vane diffuser are analyzed according to the existing ideas about the physics of the processes of energy conversion in a vane diffuser.
35
Belas, Oleg, and Andrii Belas. "General methods of forecasting nonlinear nonstationary processes based on mathematical models using statistical data." System research and information technologies, no. 1 (July 11, 2021): 79–86. http://dx.doi.org/10.20535/srit.2308-8893.2021.1.06.
Full textAbstract:
The article considers the problem of forecasting nonlinear nonstationary processes, presented in the form of time series, which can describe the dynamics of processes in both technical and economic systems. The general technique of analysis of such data and construction of corresponding mathematical models based on autoregressive models and recurrent neural networks is described in detail. The technique is applied on practical examples while performing the comparative analysis of models of forecasting of quantity of channels of service of cellular subscribers for a given station and revealing advantages and disadvantages of each method. The need to improve the existing methodology and develop a new approach is formulated.
36
Asghari, Morteza, Amir Dashti, Mashallah Rezakazemi, Ebrahim Jokar, and Hadi Halakoei. "Application of neural networks in membrane separation." Reviews in Chemical Engineering 36, no. 2 (January 28, 2020): 265–310. http://dx.doi.org/10.1515/revce-2018-0011.
Full textAbstract:
AbstractArtificial neural networks (ANNs) as a powerful technique for solving complicated problems in membrane separation processes have been employed in a wide range of chemical engineering applications. ANNs can be used in the modeling of different processes more easily than other modeling methods. Besides that, the computing time in the design of a membrane separation plant is shorter compared to many mass transfer models. The membrane separation field requires an alternative model that can work alone or in parallel with theoretical or numerical types, which can be quicker and, many a time, much more reliable. They are helpful in cases when scientists do not thoroughly know the physical and chemical rules that govern systems. In ANN modeling, there is no requirement for a deep knowledge of the processes and mathematical equations that govern them. Neural networks are commonly used for the estimation of membrane performance characteristics such as the permeate flux and rejection over the entire range of the process variables, such as pressure, solute concentration, temperature, superficial flow velocity, etc. This review investigates the important aspects of ANNs such as methods of development and training, and modeling strategies in correlation with different types of applications [microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), reverse osmosis (RO), electrodialysis (ED), etc.]. It also deals with particular types of ANNs that have been confirmed to be effective in practical applications and points out the advantages and disadvantages of using them. The combination of ANN with accurate model predictions and a mechanistic model with less accurate predictions that render physical and chemical laws can provide a thorough understanding of a process.
37
Schwartz, Gustavo Ariel. "Prediction of Rheometric Properties of Compounds by Using Artificial Neural Networks." Rubber Chemistry and Technology 74, no. 1 (March 1, 2001): 116–23. http://dx.doi.org/10.5254/1.3547632.
Full textAbstract:
Abstract The ability of an Artificial Neural Network (ANN) to evaluate the variability of rheometric properties of rubber compounds from their formulation is presented. Because of the complexity and non-linearity of mixing processes, an exact mathematical treatment of the problem is extremely difficult, or even impossible. The use of artificial neural networks (ANNs) might be very useful to analyze these processes, since they have the ability to map nonlinear relationships without prior information about process or system models. In this work a three-layer ANN is used and the optimum parameters are determined. The results are compared with theoretical and experimental published data. The dependence of the rheometric properties as a function of compound components is also analyzed. Finally, the sensibility matrix concept is introduced. The sensibility matrix allows us to calculate the minimum expected variability, for a given compound, due to the weight tolerances of its components.
38
Movellan, Javier R., Paul Mineiro, and R. J. Williams. "A Monte Carlo EM Approach for Partially Observable Diffusion Processes: Theory and Applications to Neural Networks." Neural Computation 14, no. 7 (July 1, 2002): 1507–44. http://dx.doi.org/10.1162/08997660260028593.
Full textAbstract:
We present a Monte Carlo approach for training partially observable diffusion processes. We apply the approach to diffusion networks, a stochastic version of continuous recurrent neural networks. The approach is aimed at learning probability distributions of continuous paths, not just expected values. Interestingly, the relevant activation statistics used by the learning rule presented here are inner products in the Hilbert space of square integrable functions. These inner products can be computed using Hebbian operations and do not require backpropagation of error signals. Moreover, standard kernel methods could potentially be applied to compute such inner products. We propose that the main reason that recurrent neural networks have not worked well in engineering applications (e.g., speech recognition) is that they implicitly rely on a very simplistic likelihood model. The diffusion network approach proposed here is much richer and may open new avenues for applications of recurrent neural networks. We present some analysis and simulations to support this view. Very encouraging results were obtained on a visual speech recognition task in which neural networks outperformed hidden Markov models.
39
Maass, Wolfgang. "Fast Sigmoidal Networks via Spiking Neurons." Neural Computation 9, no. 2 (February 1, 1997): 279–304. http://dx.doi.org/10.1162/neco.1997.9.2.279.
Full textAbstract:
We show that networks of relatively realistic mathematical models for biological neurons in principle can simulate arbitrary feedforward sigmoidal neural nets in a way that has previously not been considered. This new approach is based on temporal coding by single spikes (respectively by the timing of synchronous firing in pools of neurons) rather than on the traditional interpretation of analog variables in terms of firing rates. The resulting new simulation is substantially faster and hence more consistent with experimental results about the maximal speed of information processing in cortical neural systems. As a consequence we can show that networks of noisy spiking neurons are “universal approximators” in the sense that they can approximate with regard to temporal coding any given continuous function of several variables. This result holds for a fairly large class of schemes for coding analog variables by firing times of spiking neurons. This new proposal for the possible organization of computations in networks of spiking neurons systems has some interesting consequences for the type of learning rules that would be needed to explain the self-organization of such networks. Finally, the fast and noise-robust implementation of sigmoidal neural nets by temporal coding points to possible new ways of implementing feedforward and recurrent sigmoidal neural nets with pulse stream VLSI.
40
Posner, Joseph, Vivian Dickens, Andrew DeMarco, Sarah Snider, Peter Turkeltaub, and Rhonda Friedman. "4488 Neural Network of the Cognitive Model of Reading." Journal of Clinical and Translational Science 4, s1 (June 2020): 140–41. http://dx.doi.org/10.1017/cts.2020.415.
Full textAbstract:
OBJECTIVES/GOALS: A particularly debilitating consequence of stroke is alexia, an acquired impairment in reading. Cognitive models aim to characterize how information is processed based on behavioral data. If we can concurrently characterize how neural networks process that information, we can enhance the models to reflect the neuronal interactions that drive them. METHODS/STUDY POPULATION: There will be 10 unimpaired adult readers. Two functional localizer tasks, deigned to consistently activate robust language areas, identify the regions of interest that process the cognitive reading functions (orthography, phonology, semantics). Another task, designed for this experiment, analyses the reading-related functional-connectivity between these areas by presenting words classified along the attributes of frequency, concreteness, and regularity, which utilize specific cognitive routes, and a visual control. Connectivity is analyzed during word reading overall vs. a control condition to determine overall reading-related connectivity, and while reading words that have high vs. low attribute values, to determine if cognitive processing routes bias the neural reading network connectivity. RESULTS/ANTICIPATED RESULTS: The localizer analysis is expected to result in the activation of canonical reading areas. The degree of functional connectivity observed between these regions is expected to depend on the degree to which each cognitive route is utilized to read a given word. After orthographic, phonologic, and semantic areas have been identified, the connectivity analysis should show that there is high correlation between all three types of areas during reading compared to the control condition. Then the frequency, regularity, and concreteness of the words being read should alter the reliance on the pathways between these area types. This would support the hypothesized pattern of connectivity as predicted by the cognitive reading routes. Otherwise, it will show how the neural reading network differs from the cognitive model. DISCUSSION/SIGNIFICANCE OF IMPACT: The results will determine the relationship between the cognitive reading model and the neural reading network. Cognitive models show what processes occur in the brain, but neural networks show how these processes occur. By relating these components, we obtain a more complete view of reading in the brain, which can inform future alexia treatments.
41
Maass, Wolfgang, and Eduardo D. Sontag. "Neural Systems as Nonlinear Filters." Neural Computation 12, no. 8 (August 1, 2000): 1743–72. http://dx.doi.org/10.1162/089976600300015123.
Full textAbstract:
Experimental data show that biological synapses behave quite differently from the symbolic synapses in all common artificial neural network models. Biological synapses are dynamic; their “weight” changes on a short timescale by several hundred percent in dependence of the past input to the synapse. In this article we address the question how this inherent synaptic dynamics (which should not be confused with long term learning) affects the computational power of a neural network. In particular, we analyze computations on temporal and spatiotemporal patterns, and we give a complete mathematical characterization of all filters that can be approximated by feedforward neural networks with dynamic synapses. It turns out that even with just a single hidden layer, such networks can approximate a very rich class of nonlinear filters: all filters that can be characterized by Volterra series. This result is robust with regard to various changes in the model for synaptic dynamics. Our characterization result provides for all nonlinear filters that are approximable by Volterra series a new complexity hierarchy related to the cost of implementing such filters in neural systems.
42
Soen, Alexander, Alexander Mathews, Daniel Grixti-Cheng, and Lexing Xie. "UNIPoint: Universally Approximating Point Processes Intensities." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 11 (May 18, 2021): 9685–94. http://dx.doi.org/10.1609/aaai.v35i11.17165.
Full textAbstract:
Point processes are a useful mathematical tool for describing events over time, and so there are many recent approaches for representing and learning them. One notable open question is how to precisely describe the flexibility of point process models and whether there exists a general model that can represent all point processes. Our work bridges this gap. Focusing on the widely used event intensity function representation of point processes, we provide a proof that a class of learnable functions can universally approximate any valid intensity function. The proof connects the well known Stone-Weierstrass Theorem for function approximation, the uniform density of non-negative continuous functions using a transfer functions, the formulation of the parameters of a piece-wise continuous functions as a dynamic system, and a recurrent neural network implementation for capturing the dynamics. Using these insights, we design and implement UNIPoint, a novel neural point process model, using recurrent neural networks to parameterise sums of basis function upon each event. Evaluations on synthetic and real world datasets show that this simpler representation performs better than Hawkes process variants and more complex neural network-based approaches. We expect this result will provide a practical basis for selecting and tuning models, as well as furthering theoretical work on representational complexity and learnability.
43
Kuzmenko, O., H. Yarovenko, and L. Skrynka. "ANALYSIS OF MATHEMATICAL MODELS FOR COUNTERING CYBER FRAUD IN BANKS." Vìsnik Sumsʹkogo deržavnogo unìversitetu 2022, no. 2 (2022): 111–20. http://dx.doi.org/10.21272/1817-9215.2022.2-13.
Full textAbstract:
The article is devoted to the current topic of analysis of mathematical models for countering cyber fraud in banks. This problem is due to the security risks growth in the banking system, which are formed by fraudsters' cyberattacks and cybercrimes implementation. Therefore, the priority task for cyberbanking security is the application of modern mathematical methods to analyse the sources of cyber attacks, identify threats and losses in the banking services market, identify cyber-attacks and assess the scenario of potential cyber risk, etc. The article analyses the most widespread types of cyber fraud: social engineering, phishing, stalking, farming, DoS attacks, online fraud, potentially unwanted programs, etc. The study also considered a model of cognitive computing and detection of suspicious transactions in banking cyber-physical systems based on quantum computing in BCPS for the post-quantum era. The advantages, disadvantages and results of the model are defined. Predictive modelling is proposed to detect fraud in real-time by analysing incoming bank transactions with payment cards. Within the framework of this method, such models are used for the classification of fraud detection as logistic regression, a decision tree, and a narrower technique - a random forest decision tree. The study also considered using the harmonic search algorithm in neural networks to improve fraud detection in the banking system. It is found that although this model has the advantage of learning ability based on past behaviour, there are difficulties in the long-term processing of many neural networks. The stages of model implementation are also given. In addition, the modelling of credit card fraud detection is based on using two types of models: supervised and unsupervised. Supervised models include logistic regression, K-nearest neighbours, and extreme gradient boosting. The one-class support vector model, restricted Boltzmann model, and generative-competitive network are considered among uncontrolled generative models.
44
Mostafa, Hesham, and Gert Cauwenberghs. "A Learning Framework for Winner-Take-All Networks with Stochastic Synapses." Neural Computation 30, no. 6 (June 2018): 1542–72. http://dx.doi.org/10.1162/neco_a_01080.
Full textAbstract:
Many recent generative models make use of neural networks to transform the probability distribution of a simple low-dimensional noise process into the complex distribution of the data. This raises the question of whether biological networks operate along similar principles to implement a probabilistic model of the environment through transformations of intrinsic noise processes. The intrinsic neural and synaptic noise processes in biological networks, however, are quite different from the noise processes used in current abstract generative networks. This, together with the discrete nature of spikes and local circuit interactions among the neurons, raises several difficulties when using recent generative modeling frameworks to train biologically motivated models. In this letter, we show that a biologically motivated model based on multilayer winner-take-all circuits and stochastic synapses admits an approximate analytical description. This allows us to use the proposed networks in a variational learning setting where stochastic backpropagation is used to optimize a lower bound on the data log likelihood, thereby learning a generative model of the data. We illustrate the generality of the proposed networks and learning technique by using them in a structured output prediction task and a semisupervised learning task. Our results extend the domain of application of modern stochastic network architectures to networks where synaptic transmission failure is the principal noise mechanism.
45
Panasenko, Natalia, Nikolay Motuz, and Asya Atayan. "Assimilation and processing of observation data obtained by satellite earth sensing for monitoring the current state of heterogeneous objects on the water surface." E3S Web of Conferences 224 (2020): 02030. http://dx.doi.org/10.1051/e3sconf/202022402030.
Full textAbstract:
The study is devoted to the analysis of satellite observations data assimilation to discover the necessary information for developing and verifying mathematical models of hydrodynamics and biological shallowwater kinetics. The use of satellite earth sensing data is taken to enhance information base. The possible use of neural networks with optical flow computation is considered in the study. The objective of the study is to develop a software tool used to identify the initial conditions in mathematical modeling of hydrobilogical shallow-water processes.
46
DiMattina, Christopher, and Kechen Zhang. "Active Data Collection for Efficient Estimation and Comparison of Nonlinear Neural Models." Neural Computation 23, no. 9 (September 2011): 2242–88. http://dx.doi.org/10.1162/neco_a_00167.
Full textAbstract:
The stimulus-response relationship of many sensory neurons is nonlinear, but fully quantifying this relationship by a complex nonlinear model may require too much data to be experimentally tractable. Here we present a theoretical study of a general two-stage computational method that may help to significantly reduce the number of stimuli needed to obtain an accurate mathematical description of nonlinear neural responses. Our method of active data collection first adaptively generates stimuli that are optimal for estimating the parameters of competing nonlinear models and then uses these estimates to generate stimuli online that are optimal for discriminating these models. We applied our method to simple hierarchical circuit models, including nonlinear networks built on the spatiotemporal or spectral-temporal receptive fields, and confirmed that collecting data using our two-stage adaptive algorithm was far more effective for estimating and comparing competing nonlinear sensory processing models than standard nonadaptive methods using random stimuli.
47
Head, R., D. Shepherd, G. Butt, and G. Buck. "OTTER mathematical process simulation of potable water treatment." Water Supply 2, no. 1 (January 1, 2002): 95–101. http://dx.doi.org/10.2166/ws.2002.0012.
Full textAbstract:
Process modelling has been used for many years in the chemical engineering field and more recently has become well established for as a tool for analysing and optimising the performance of wastewater treatment works. In the clean water area, models are routinely used for simulating distribution networks. In contrast, however, the use of modelling tools on potable water treatment works is relatively new and has yet to become well established. A range of tools have been suggested, including artificial neural networks, computational fluid dynamics and process simulation. WRc have developed a dynamic simulation package for predicting the performance of water treatment works, via models of individual processes. The software has a range of uses, including process and works optimisation, operational decision support, as a design aid and for training engineers and operators. The models are dynamic so that they predict the response of the treatment works to changes in flow, raw water quality and process operating conditions. The software has been used in a wide variety of applications, including optimising process plant operation to minimise cost and to investigate the reasons why a treatment works failed to meet its design criteria at the maximum design throughput.
48
Nerella, Santhi Sree, Sudheer V. V. S. Nakka, and Bhramara Panitapu. "Mathematical Modeling of Closed Loop Pulsating Heat Pipe by Using Artificial Neural Networks." International Journal of Heat and Technology 39, no. 3 (June 30, 2021): 955–62. http://dx.doi.org/10.18280/ijht.390332.
Full textAbstract:
Pulsating heat pipe is one of the prominent technology for thermal management of electronic devices. It consists of three sections namely evaporator, adiabatic and condenser section. PHP is a two phase passive device having efficient and quick ability of transferring heat from evaporator section to condenser section. At first an 8 turn pulsating heat pipe of closed loop ends (CLPHP) with copper tube capillary dimensions is investigated experimentally for different fill ratios and for different inclinations by varying range of heat inputs. Different working fluids viz Water, Acetone, Ethanol and Methanol are considered for the experimentation. One of the recent analytical technology for modelling of CLPHPs is Artificial Neural Network (ANN) approach. The analytical models are having limited scope of applicability and they are simple in nature. The present paper describes Validation of experimental data by training prediction model ANN with available data. Three input nodes such as input heat, fill ratio and angle of inclination and one output node corresponding to PHP that is thermal resistance are considered. The feed forward neural network (FFNN) architecture is adopted for predictions. By using the physical phenomena of the system modelling are clearly known for obtaining feasible results which is main function of ANN. The predicted data validates experimental data in a satisfactory range and the results are found to be in good agreement with in the range of ± 10 percent.
49
Шаповалова, И. А. "TIME DELAY NEURAL NETWORK MODELING IMMUNE SYSTEM." Южно-Сибирский научный вестник, no. 3(37) (June 30, 2021): 43–48. http://dx.doi.org/10.25699/sssb.2021.37.3.016.
Full textAbstract:
Современная иммунология не может успешно развиваться без помощи математического моделирования. Математические модели являются эффективным фильтром идей и индикатором правильности выбранных предположений, позволяют дать правильную интерпретацию результатам и выбирать критерии для оценки правильности, могут быть использованы как средство для визуализации результатов вычисления, что помогает дальнейшему развитию вычислительных алгоритмов. Исследование математической модели иммунной системы позволяет сравнивать теоретические и экспериментальные результаты и уточнять предположения, положенные в основу математического моделирования. Иммунная система является высокоразвитой биологической системой, функция которой заключается в выявлении и уничтожении чужеродного агента, поэтому она должна распознавать разнообразных возбудителей. Иммунная система способна к обучению, запоминанию, распознаванию образов, аналогичными свойствами обладают искусственные нейронные сети. Искусственные нейронные сети, подобно биологическим, являются вычислительной системой с огромным числом параллельно функционирующих простых процессоров с огромным числом связей. Нейросетевые алгоритмы используются в кластеризации, визуализации данных, контроле и оптимизации управляемых процессов, разработке искусственных нейронных сетей. В работе исследуется математическая модель иммунной системы, которая моделируется с помощью искусственной нейронной сети и описывается системой дифференциальных уравнений с запаздыванием. При анализе модели используется аппарат математической теории оптимального управления, а именно принцип максимума для систем дифференциальных уравнений с запаздыванием в аргументе функции состояния и аппарат методов оптимизации, базирующийся на методе быстрого автоматического дифференцирования. Вместо традиционных методов программирования используется обучение полносвязной искусственной нейронной сети с помощью метода распространения ошибки. Modern immunology can not be developed successfully without the help of mathematical modeling. Mathematical models are an effective way filter and indicator of the correctness of the selected assumptions. Mathematical models allow us to give a correct interpretation of the results, to select criteria for evaluating the correctness and that help the development of the numerical methods and algorithm. The research of the mathematical model of the immune system allow to compare theoretical and experimental results and clarified mathematical assumptions laid down in the basis of mathematical modeling. The immune system is a highly developed biological system, whose function is to detect and destroy foreign substance, so it needs to recognize a variety of pathogens.The immune system is capable of learning to remember the recognitions of images. The similar properties possess artificial neural networks. Similar to biological ones artificial neural networks are computer systems with a huge number of parallel functioning simple processors and with a large number of connections. Neural networks algorithms are used in clustering, data visualization, control and optimization of processes, the development of artificial neural networks. In the article we consider mathematical model of immune system modeled with the help of artificial multi layer neural net described by the system of differential equations with delay in argument of state functions. The model is analyzed with the help of the theory of optimal control namely the maximum principle of Pontrjagin for the systems of differential equations with delay in argument of the state functions. The method of optimization is based on the method of fast automatic differentiations. Instead of traditional methods of programming the training of the fully connected neural networks and the error propagation method are used.
50
Tomasevic, Nikola M., Aleksandar M. Neskovic, and Natasa J. Neskovic. "Correlated EEG Signals Simulation Based on Artificial Neural Networks." International Journal of Neural Systems 27, no. 05 (May 3, 2017): 1750008. http://dx.doi.org/10.1142/s0129065717500083.
Full textAbstract:
In recent years, simulation of the human electroencephalogram (EEG) data found its important role in medical domain and neuropsychology. In this paper, a novel approach to simulation of two cross-correlated EEG signals is proposed. The proposed method is based on the principles of artificial neural networks (ANN). Contrary to the existing EEG data simulators, the ANN-based approach was leveraged solely on the experimentally acquired EEG data. More precisely, measured EEG data were utilized to optimize the simulator which consisted of two ANN models (each model responsible for generation of one EEG sequence). In order to acquire the EEG recordings, the measurement campaign was carried out on a healthy awake adult having no cognitive, physical or mental load. For the evaluation of the proposed approach, comprehensive quantitative and qualitative statistical analysis was performed considering probability distribution, correlation properties and spectral characteristics of generated EEG processes. The obtained results clearly indicated the satisfactory agreement with the measurement data.