Academic literature on the topic 'Adaptation à la vie extra-Utérine'

The Big Sleep: Marlowe and Vivian practising kissing; General Sternwood shivering in a hothouse full of orchids; a screenplay, co-written by Faulkner, famously mysterious and difficult to solve. Released in 1946, Howard Hawks' adaptation of Raymond Chandler reunited Bogart and Bacall and gave them two of their most famous roles. The mercurial but ever-manipulative Hawks dredged humour and happiness out of film noir. 'Give him a story about more murders than anyone can keep up with, or explain,' David Thomson writes in his compelling study of the film, 'and somehow he made a paradise.' When it

New objects are continuously emerging in the dynamically changing world and a real-world artificial intelligence system should be capable of continual and effectual adaptation to new emerging classes without forgetting old ones. In view of this, in this paper we tackle a challenging and practical continual learning scenario named few-shot class-incremental learning (FSCIL), in which labeled data are given for classes in a base session but very limited labeled instances are available for new incremental classes. To address this problem, we propose a novel and succinct approach by introducing deep dictionary learning which is a hybrid learning architecture that combines dictionary learning and visual representation learning to provide a better space for characterizing different classes. We simultaneously optimize the dictionary and the feature extraction backbone in the base session, but only finetune the dictionary in the incremental session for adaptation to novel classes, which can alleviate the forgetting problem on base classes compared to finetuning the entire model. To further facilitate future adaptation, we also incorporate multiple pseudo classes into the base session training so that certain space projected by dictionary can be reserved for future new concepts. Moreover, we adopt a novel optimization process by implicitly incorporating the dictionary reconstruction loss into the learning process without adding extra loss terms. The extensive experimental results on CIFAR100, miniImageNet and CUB200 validate the effectiveness of our approach compared to other SOTA methods.

Abstract This chapter considers the role of cooking in the evolution of human diet. People in every culture know how to make fire, and everywhere they use it to improve their food (Tylor, 1878; Gott, 2002; Wrangham and Conklin-Brittain, 2003; fig. 16.1). But “no beast is a cook,” as Boswell (1773) asserted. This difference between humans and other animals has long been appreciated, and some have even used it to define us: Boswell (1773) called humans the “cooking animal.” But while cooking is indisputably unique to humans, its evolutionary significance is a matter of debate. There are two contrasting views. The first, which is conventional wisdom, sees cooking as merely one of many extra-oral food-processing techniques (such as pounding, grinding, or drying) that can raise food quality. According to this view, cooking may be valuable in facilitating meal preparation but it is not important for understanding human adaptation. For example, it would not be considered to have led to fundamental changes in the human digestive system. In line with this idea, the ecological effect of human diet choice and foraging strategies is often discussed without considering the influence of cooking (e.g., Kaplan et al., 2000).

Pre-trained Large Language Models (LLMs) have demonstrated prominent generalization to various linguistic tasks. However, due to the inherent modality and task discrepancy, parameter-efficient transfer learning for adapting LLMs to vision-language (VL) tasks remains challenging, which may struggle with excessive extra computation and data expenditure for VL pre-training and disconnection between multi-modal representations. This paper concentrates on the parameter-efficient adaptation of LLMs to VL tasks without inflexible multi-modal alignment pre-training on additional image-text pairs. Inspired by Instruction Tuning and the nature of multi-modal representation learning, we propose Multi-modal Prompt Tuning for Language Models (MPT4LM). This method provides text-relevant visual prompts via a plug-and-play Cross-Attention module and integrates them with textual Learnable Instruction as multi-modal prompts into LLMs. We further assemble MPT4LM with the currently prevalent Adapter approach to alleviate the trainable parameter scale and facilitate the collaboration of multi-modal prompts. We evaluate MPT4LM upon two representative LLMs: LLAMA-2 and Flan-T5, over two VL tasks: Visual Question Answering (VQAv2.0, GQA) and Visual Entailment (SNLI-VE). Extensive experimental results reveal that MPT4LM achieves state-of-the-art performance among prompting methods with only fine-tuning about 0.65% of the parameters of backbones, indicating a better trade-off between computation and data overhead and model performance. Our code is available at: https://github.com/YzM1a0/MPT4LM.

Abstract At an ever-faster pace, market’s requirements for operational flexibility and turn-down capabilities are becoming the fundamental driver for heavy-duty gas turbine design and upgrade. This enables better compensation of the natural intermittence of renewable energy sources such as wind and solar by effectively compensating for both surpluses and shortages compared to the electric grid power request. Gas turbine-based energy plants require a lowest Minimum Environmental Load (MEL) for times when the power request is mainly covered by renewable sources and, however, need to promptly res