Adversarial Variational Optimization of Non-Differentiable Simulators

Abstract

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 generative adversarial networks, variational optimization and empirical Bayes. 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 JS divergence between the marginal distribution of the synthetic data and the empirical distribution of observed data is minimized. We evaluate and compare the method with simulators producing both discrete and continuous data.

v4: Final version published at AISTATS 2019; v5: Fixed typo in Eqn 13

References in corpus (10)

• Improved Training of Wasserstein GANs
• Which Training Methods for GANs do actually Converge?
• Towards Principled Methods for Training Generative Adversarial Networks
• Variational Approaches for Auto-Encoding Generative Adversarial Networks
• Learning in Implicit Generative Models
• The Numerics of GANs
• Hierarchical Implicit Models and Likelihood-Free Variational Inference
• Variational Inference using Implicit Distributions
• ELFI: Engine for Likelihood-Free Inference
• Variational Inference over Non-differentiable Cardiac Simulators using Bayesian Optimization

Cited by in corpus (11)

• Machine learning and the physical sciences
• Mining gold from implicit models to improve likelihood-free inference
• Mind Mappings: Enabling Efficient Algorithm-Accelerator Mapping Space Search
• Mining for Dark Matter Substructure: Inferring subhalo population properties from strong lenses with machine learning
• Likelihood-free MCMC with Amortized Approximate Ratio Estimators
• Effective LHC measurements with matrix elements and machine learning
• Likelihood-free inference with an improved cross-entropy estimator
• Reframing Jet Physics with New Computational Methods
• Learning To Simulate
• Recurrent machines for likelihood-free inference
• Adaptive Divergence for Rapid Adversarial Optimization