Adversarial Variational Optimization of Non-Differentiable Simulators

07/22/2017
by   Gilles Louppe, et al.
0

Complex computer simulators are increasingly used across fields of science as generative models tying parameters of an underlying theory to experimental observations. Inference in this setup is often difficult, as simulators rarely admit a tractable density or likelihood function. We introduce Adversarial Variational Optimization (AVO), a likelihood-free inference algorithm for fitting a non-differentiable generative model incorporating ideas from empirical Bayes and variational inference. We adapt the training procedure of generative adversarial networks by replacing the differentiable generative network with a domain-specific simulator. We solve the resulting non-differentiable minimax problem by minimizing variational upper bounds of the two adversarial objectives. Effectively, the procedure results in learning a proposal distribution over simulator parameters, such that the corresponding marginal distribution of the generated data matches the observations. We present results of the method with simulators producing both discrete and continuous data.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/07/2018

Semi-Amortized Variational Autoencoders

Amortized variational inference (AVI) replaces instance-specific local i...
research
12/09/2017

Variational Inference over Non-differentiable Cardiac Simulators using Bayesian Optimization

Performing inference over simulators is generally intractable as their r...
research
11/30/2018

Recurrent machines for likelihood-free inference

Likelihood-free inference is concerned with the estimation of the parame...
research
06/06/2015

Approximating Likelihood Ratios with Calibrated Discriminative Classifiers

In many fields of science, generalized likelihood ratio tests are establ...
research
05/25/2016

Asymptotically exact inference in differentiable generative models

Many generative models can be expressed as a differentiable function of ...
research
07/05/2018

Learning in Variational Autoencoders with Kullback-Leibler and Renyi Integral Bounds

In this paper we propose two novel bounds for the log-likelihood based o...
research
12/21/2017

Truncated Variational Sampling for "Black Box" Optimization of Generative Models

We investigate the optimization of two generative models with binary hid...

Please sign up or login with your details

Forgot password? Click here to reset