Master AI and Machine Learning: From Neural Networks to Applications
Learn AI, Data Science & Business — Earn Certificates That Get You Hired
Overview
Google, IBM & Meta Certificates – 40% Off
One Coursera Plus subscription covers most Professional Certificates on Coursera.
Unlock All Certificates
This advanced lecture examines variational autoencoders for natural language processing, covering generative and discriminative models, evidence lower bounds, training strategies, discrete latent variables, and sentence generation.
Syllabus
Introduction
Discriminative vs generative models
Types of variables
Loss function
Two tasks
Bias and variance
Evidence lower bound
Procedural training
Questions
Learning the VAE
Generating Sentences
Problems
kl divergence annealing
Free bits
Weaken the decoder
Aggressive inference network learning
Standard variational autoencoder
What are discrete latent variables
Method 1 Sampling
Method 2 Sampling
Method 2 Reparameterization
Taught by
Graham Neubig