• Stars
    star
    349
  • Rank 121,528 (Top 3 %)
  • Language
    Python
  • Created over 8 years ago
  • Updated about 6 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

code for Structured Variational Autoencoders

Code 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

More Repositories

1

autodidact

A pedagogical implementation of Autograd
Jupyter Notebook
945
star
2

pyhsmm

Python
548
star
3

pybasicbayes

Python
153
star
4

pyslds

Python
88
star
5

my-oh-my-zsh

Shell
84
star
6

pylds

some tools for gaussian linear dynamical systems
Python
83
star
7

pyhsmm-autoregressive

autoregressive plugin
Python
27
star
8

pyhsmm-factorial

Python
18
star
9

todo-bash

super-simple command-line todo tracking
Shell
13
star
10

ode-diff-notes

Python
12
star
11

py-manipulate

Python
10
star
12

probprog

simple probabilistic programming in Scheme
Scheme
10
star
13

matlab-hsmm

Objective-C
9
star
14

variational_autoencoder

Python
8
star
15

pplham

Python
7
star
16

spectral_clustering

Python
6
star
17

py4sid

subspace identification for linear systems
Python
6
star
18

py-diskmemo

Python
6
star
19

pykalmanfilters

Python
5
star
20

kalman_grads

Python
4
star
21

pyparticlefilters

Python
4
star
22

pyhsmm-collapsedinfinite

Python
4
star
23

config-vim

my .vim
Vim Script
4
star
24

gaussian-hogwild-gibbs

Python
4
star
25

dirichlet-truncated-multinomial

Python
4
star
26

information-theory-tutorial

Python
4
star
27

config-fish

my fish configuration
Shell
4
star
28

next.ml

JavaScript
3
star
29

yaldapy

yet another LDA implementation
Python
3
star
30

pyhsmm-beamsampling

beam sampling for pyhsmm
Python
3
star
31

pymattutil

a few miscellaneous functions I've found useful
Python
3
star
32

matlab-fftconv

MATLAB
3
star
33

autograd_linalg

improve scipy.linalg and corresponding autograd gradfuns
Python
3
star
34

cs281_linear_regression

Python
2
star
35

mattjj.github.com

2
star
36

cachedproperties

an experiment in property caching
Python
2
star
37

pyhsmm-subhmms

Python
2
star
38

dotfiles

my config files, especially .tmux.conf
Shell
2
star
39

py-stateplots

Python
2
star
40

vim-prefs

my vim prefs, separated so that it can be a bundle
Vim Script
2
star
41

mpi4py-paralleltempering

Python
1
star
42

pykinematics

Python
1
star
43

jsmattutil

JavaScript
1
star