Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a 27-minute video analyzing the groundbreaking paper "Attention Is All You Need" by Vaswani et al. Dive into the revolutionary Transformer architecture, which relies solely on attention mechanisms, eliminating the need for recurrent and convolutional neural networks. Learn how this innovative approach achieves superior performance in machine translation tasks while significantly reducing training time. Discover the key components of the Transformer, including positional encoding and the attention mechanism. Understand how this architecture generalizes well to other tasks, such as English constituency parsing. Gain insights into the paper's impact on the field of natural language processing and its potential applications in various domains.
Syllabus
Introduction
Traditional Language Processing
Attention
Longrange dependencies
Attention mechanism
Encoding
Positional Encoding
Tension
Top Right
Attention Computed
Conclusion
Taught by
Yannic Kilcher