Sequence to Sequence Models: Overview and Attention Mechanisms

This lecture covers the basics of sequence to sequence models, focusing on the shortcomings related to long-range dependencies and the temporal bottleneck. It introduces improvements through attention mechanisms, explaining how they address the limitations of traditional recurrent models. The instructor discusses the concept of bidirectional encoders and how they enhance the representation of input sequences. The training process for encoder-decoder models is explained, emphasizing the need for paired data across languages for tasks like machine translation. The lecture also delves into the implementation of attentive encoder-decoder models, highlighting the importance of attention in mitigating the temporal bottleneck and improving interpretability. Various attention functions and their role in enhancing model performance are explored, along with the interpretability aspect of attention mechanisms.

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related lectures (41)

Related concepts (35)

CS-552: Modern natural language processing

Natural language processing is ubiquitous in modern intelligent technologies, serving as a foundation for language translators, virtual assistants, search engines, and many more. In this course, stude

Statistical machine translation

Statistical machine translation (SMT) was a machine translation approach, that superseded the previous, rule-based approach because it required explicit description of each and every linguistic rule, which was costly, and which often did not generalize to other languages. Since 2003, the statistical approach itself has been gradually superseded by the deep learning-based neural network approach. The first ideas of statistical machine translation were introduced by Warren Weaver in 1949, including the ideas of applying Claude Shannon's information theory.

Recurrent neural network

A recurrent neural network (RNN) is one of the two broad types of artificial neural network, characterized by direction of the flow of information between its layers. In contrast to uni-directional feedforward neural network, it is a bi-directional artificial neural network, meaning that it allows the output from some nodes to affect subsequent input to the same nodes. Their ability to use internal state (memory) to process arbitrary sequences of inputs makes them applicable to tasks such as unsegmented, connected handwriting recognition or speech recognition.

Example-based machine translation

Example-based machine translation (EBMT) is a method of machine translation often characterized by its use of a bilingual corpus with parallel texts as its main knowledge base at run-time. It is essentially a translation by analogy and can be viewed as an implementation of a case-based reasoning approach to machine learning. At the foundation of example-based machine translation is the idea of translation by analogy.

Long short-term memory

Long short-term memory (LSTM) network is a recurrent neural network (RNN), aimed to deal with the vanishing gradient problem present in traditional RNNs. Its relative insensitivity to gap length is its advantage over other RNNs, hidden Markov models and other sequence learning methods. It aims to provide a short-term memory for RNN that can last thousands of timesteps, thus "long short-term memory".

Convolutional neural network

Convolutional neural network (CNN) is a regularized type of feed-forward neural network that learns feature engineering by itself via filters (or kernel) optimization. Vanishing gradients and exploding gradients, seen during backpropagation in earlier neural networks, are prevented by using regularized weights over fewer connections. For example, for each neuron in the fully-connected layer 10,000 weights would be required for processing an image sized 100 × 100 pixels.

Deep Learning for NLP CS-431: Introduction to natural language processing

Introduces deep learning concepts for NLP, covering word embeddings, RNNs, and Transformers, emphasizing self-attention and multi-headed attention.

Deep Learning for NLP

Delves into Deep Learning for Natural Language Processing, exploring Neural Word Embeddings, Recurrent Neural Networks, and Attentive Neural Modeling with Transformers.

Machine Translation: Sequence-to-Sequence and Attention EE-608: Deep Learning For Natural Language Processing

Explores the advancements in Machine Translation, focusing on Sequence-to-Sequence models and Attention mechanisms.

Language Modelling and Recurrent Neural Networks EE-608: Deep Learning For Natural Language Processing

Explores language modelling, RNNs, n-gram models, LSTMs, and bidirectional RNNs.

Trajectory Forecasting in Autonomous Vehicles CIVIL-459: Deep learning for autonomous vehicles

Explores trajectory forecasting in autonomous vehicles, focusing on deep learning models for predicting human trajectories in socially-aware transportation scenarios.