Neural Nets for NLP 2017 - A Simple Exercise - Predicting the Next Word in a Sentence

Explore a comprehensive lecture on neural networks for natural language processing, focusing on predicting the next word in a sentence. Delve into topics such as describing words by their context, counting and prediction techniques, skip-grams and Continuous Bag of Words (CBOW) models, and methods for evaluating and visualizing word vectors. Learn about advanced techniques for word vector creation and gain practical insights through provided slides and code examples. This lecture, part of CMU's CS 11-747 course, offers a deep dive into the fundamentals of word embeddings and their applications in NLP tasks.

Introduction
Evaluation
Language Models
Feature Eyes Models
Example
Converting Scores to Probabilities
Computation Graph
Lookup
Loss Function
Perimeter Update
Unknown Words
Vocabulary
Unfair Advantage
Problems of Previous Model
Neural Language Models
Code
InputOutput Embedding
Training Tricks
Learning Rate Decay