mattjj/svae
mattjj/svae has 354 GitHub stars and 89 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-18 01:49 UTC. Last push 2018-08-22.
Measured star growth
- in 24h
- 0
- in 7 days
- Collecting history
- in 15 days
- Collecting history
- in 1 month
- Collecting history
Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.
About
code for Structured Variational Autoencoders
Developer ToolsREADME.md
OriginalCode for Composing graphical models with neural networks for structured representations and fast inference, a.k.a. structured variational autoencoders.
NOTE: This code isn't yet compatible with a recent rewrite of autograd. To use an older, compatible version of autograd, clone autograd and check out commit 0f026ab.
###Abstract
We propose a general modeling and inference framework that composes probabilistic graphical models with deep learning methods and combines their respective strengths. Our model family augments graphical structure in latent variables with neural network observation models. For inference we extend variational autoencoders to use graphical model approximating distributions, paired with recognition networks that output conjugate potentials. All components of these models are learned simultaneously with a single objective, giving a scalable algorithm that leverages stochastic variational inference, natural gradients, graphical model message passing, and the reparameterization trick. We illustrate this framework with several example models and an application to mouse behavioral phenotyping.
By