Lecture

Cognitive Maps in Rats and Men

Graph Chat

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

Description

This lecture covers the concept of cognitive maps as studied by Edward C. Tolman, focusing on spatial learning in rats and humans. It discusses the different reward systems used in experiments and their outcomes, as well as the implications of latent learning. The lecture also delves into the topics of attention, motivation, and grouping in visual intelligence, exploring mechanisms like selective attention and alignment score functions for computing attention. Furthermore, it introduces the concept of scaled dot-product attention, explaining self-attention and cross-attention, and how attention is computed using queries, keys, and values. The lecture concludes with an overview of transformers in machine learning, detailing the transformer encoder and decoder, and emphasizing the importance of positional encodings in maintaining input order.

This video is available exclusively on Mediaspace for a restricted audience. Please log in to MediaSpace to access it if you have the necessary permissions.

Watch on Mediaspace

Official source

Related concepts (22)

Transformer (machine learning model)

A transformer is a deep learning architecture that relies on the parallel multi-head attention mechanism. The modern transformer was proposed in the 2017 paper titled 'Attention Is All You Need' by Ashish Vaswani et al., Google Brain team. It is notable for requiring less training time than previous recurrent neural architectures, such as long short-term memory (LSTM), and its later variation has been prevalently adopted for training large language models on large (language) datasets, such as the Wikipedia corpus and Common Crawl, by virtue of the parallelized processing of input sequence.

Attention

Attention is the concentration of awareness on some phenomenon to the exclusion of other stimuli. It is a process of selectively concentrating on a discrete aspect of information, whether considered subjective or objective. William James (1890) wrote that "Attention is the taking possession by the mind, in clear and vivid form, of one out of what seem several simultaneously possible objects or trains of thought. Focalization, concentration, of consciousness are of its essence.

Attention (machine learning)

Machine learning-based attention is a mechanism mimicking cognitive attention. It calculates "soft" weights for each word, more precisely for its embedding, in the context window. It can do it either in parallel (such as in transformers) or sequentially (such as recursive neural networks). "Soft" weights can change during each runtime, in contrast to "hard" weights, which are (pre-)trained and fine-tuned and remain frozen afterwards. Multiple attention heads are used in transformer-based large language models.

Generative pre-trained transformer

Generative pre-trained transformers (GPT) are a type of large language model (LLM) and a prominent framework for generative artificial intelligence. The first GPT was introduced in 2018 by OpenAI. GPT models are artificial neural networks that are based on the transformer architecture, pre-trained on large data sets of unlabelled text, and able to generate novel human-like content. As of 2023, most LLMs have these characteristics and are sometimes referred to broadly as GPTs.

Neural machine translation

Neural machine translation (NMT) is an approach to machine translation that uses an artificial neural network to predict the likelihood of a sequence of words, typically modeling entire sentences in a single integrated model. They require only a fraction of the memory needed by traditional statistical machine translation (SMT) models. Furthermore, unlike conventional translation systems, all parts of the neural translation model are trained jointly (end-to-end) to maximize the translation performance.

Related lectures (35)

Generative Models: Self-Attention and Transformers

Covers generative models with a focus on self-attention and transformers, discussing sampling methods and empirical means.

Latent Semantic Indexing: Concepts and Applications

Explores Latent Semantic Indexing, a technique for mapping documents into a concept space for retrieval and classification.

From Attention to Transformers

Explores the evolution from attention mechanisms to transformers in modern NLP, emphasizing the significance of self-attention and cross-attention.

Document Retrieval and Classification

Covers document retrieval, classification, sentiment analysis, and topic detection using TF-IDF matrices and contextualized word vectors like BERT.

Transformers: Pretraining and Decoding Techniques