Giter Club home page Giter Club logo

compiled-inference's Introduction

Learn proposal distributions for importance sampling and SMC

This code provides a PyTorch implementation of the method described in this paper for training neural networks which can be used as proposal distributions for importance sampling or sequential Monte Carlo:

Paige, B., & Wood, F. (2016). Inference Networks for Sequential Monte Carlo in Graphical Models. In Proceedings of the 33rd International Conference on Machine Learning. JMLR W&CP 48: 3040-3049.

The largest section of re-usable code is an implementation of a conditional variant of MADE as a PyTorch module, found in learn_smc_proposals.cde. This can be used to fit a conditional density estimator. There is a version for real-valued data and a version for binary data.

  • The linear regression notebook provides an end-to-end usage example. This notebook defines a generative model using PyMC, a non-conjugate regression model. It then goes through the process of defining a network which will represent the inverse, training it on samples from the joint distribution, and then using it for inference.

Two more involved examples are implemented in learn_smc_proposals.examples; pre-trained weights are included in this repository. Figures and inference are shown in two notebooks:

compiled-inference's People

Contributors

tbrx avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar

Forkers

leezqcst talesa

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.