Neural Nets for NLP 2021 - Machine Reading with Neural Nets

Learn about machine reading with neural networks in this comprehensive lecture from CMU's Neural Networks for NLP course. Explore various machine reading datasets, methods for encoding context and multi-hop reasoning, and important caveats about dataset biases. Dive into topics such as attention models, span selection, question decomposition, and retrieval-based question answering. Examine real-world examples from Daily Mail and natural questions datasets, and understand the challenges of adversarial examples and symbolic reasoning in machine reading tasks. Gain insights into the latest techniques for improving neural network performance in natural language processing and question answering systems.

Introduction
Machine Reading
Multiple Choice Questions
Span Selection Tasks
Closed Questions
Why Machine Reading
Attention Models
Attention Flow
Span Selection
Refinement
Multistep reasoning
Multistep data sets
Multihop reasoning
Retrievalbased question answering
Language models
Question decomposition
Question answering with context
Data bias
Reading comprehension example
Daily Mail example
adversarial examples
adversarial data sets
natural questions
symbolic reasoning
semantic parsing
outro