Kamath Uday. Transformers for Machine Learning - A Deep Dive

CRC Press, 2022. — 283 p. — ISBN: .978-1-003-17008-2.Transformers are becoming a core part of many neural network architectures, employed in a wide range of applications such as NLP, Speech Recognition, Time Series, and Computer Vision. Transformers have gone through many adaptations and alterations, resulting in newer techniques and methods. Transformers for Machine Learning: A Deep Dive is the first comprehensive book on transformers.Key Features.
A comprehensive reference book for detailed explanations of every algorithm and technique related to the transformers.
60+ transformer architectures covered comprehensively.
A book for understanding how to apply the transformer techniques in speech, text, time series, and computer vision.
Practical tips and tricks for each architecture and how to use it in the real world.
Hands-on case studies and code snippets for theory and practical real-world analysis using the tools and libraries, all ready to run in Google Colab.The theoretical explanations of the state-of-the-art transformer architectures will appeal to postgraduate students and researchers (academic and industry) as it will provide a single entry point with deep discussions of a quickly moving field. The practical hands-on case studies and code will appeal to undergraduate students, practitioners, and professionals as it allows for quick experimentation and lowers the barrier to entry into the field.Contents.
Foreword.
Preface.
Authors.
Contributors.
1 - Deep Learning and Transformers: An Introduction.
Deep Learning: A Historic Perspective.
Transformer Sandtaxonomy.
Resources.2 - Transformers: Basics and Introduction.
Encoder-Decoder Architecture.
Sequence-To-Sequence.
Attention Mechanism.
Transformer.
Case Study: Machine Translation.3 - Bidirectional Encoder Representations from Transformers (BERT).
Bert.
Bert Variants.
Applications.
Bert Insights.
Case Study: Topic Modeling With Transformers.
Case Study: Fine-Tuning Bert.4 - Multilingual Transformer Architectures.
Multilingual Transformer Architectures.
Multilingual Data.
Multilingual Transfer Learning In Sights.
Case Study.5 - Transformer Modifications.
Transformerblockmodifications.
Transformers With Modified Multi-Head Self-Attention.
Modifications For Training Task Efficiency.
Transformer Submodule Changes.
Case Study: Sentiment Analysis.6 - Pre-trained and Application-Specific Transformers.
Text Processing.
Computer Vision.
Automatic Speech Recognition.
Multimodal And Multitasking Transformer.
Video Processing With Timesformer.
Graph Transformers.
Reinforcement Learning.
Case Study: Automatic Speech Recognition.7 - Interpretability and Explainability Techniques for Transformers.
Traits Of Explainable Systems.
Related Areas That Impact Explainability.
Explainable Methods Taxonomy.
Attention And Explanation.
Quantifying Attention Flow.
Case Study: Text Classification With Explainability.Bibliography.
Index.True PDF.