tensorflow / tensorflow/probability

Feature Request: Gumbel Mixture Models

Open
#1,598 4 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
4.4k
Forks
1.1k
PR merge metrics
No merged PRs in 30d

Description

It is possible to construct reparameterizable mixture distributions by replacing the categorical distribution with a gumbel (relaxed categorical) distribution. The ability to use a relaxed one-hot categorical distribution in mixture or mixtureSameFamily would be potentially very useful.

Differentiable mixture distributions implemented in torch here:
https://github.com/nextBillyonair/DPM/blob/master/dpm/distributions/gumbel_mixture_model.py

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing the linked reference implementation at dpm/distributions/gumbel_mixture_model.py and the existing mixture and mixtureSameFamily entry points in TensorFlow Probability. Determine the intended relaxed one-hot categorical behavior and define tests that establish reparameterization and mixture usage before considering the feature complete.

Written by the indexing model from the issue text.

Assessment

Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.