conftrace_

2011 AISTATS AISTATS 2011

A conditional game for comparing approximations

Abstract

We present a “conditional game” to be played between two approximate inference algorithms. We prove that exact inference is an optimal strategy and demonstrate how the game can be used to estimate the relative accuracy of two different approximations in the absence of exact marginals.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Mathematics & Optimization

📈 Trend Setter - Game Theory

🐣 Hot Topic Early Bird - game theory

Authors

Frederik Eaton

Topics

Artificial Intelligence > Core AI > Game AI Machine Learning > Optimization & Theory > Bayesian Inference Mathematics & Optimization > Optimization > Game Theory

Keywords

approximate inference game theory game-theoretic analysis conditional game relative accuracy estimation

More from AISTATS 2011

All AISTATS 2011 papers →

Discussion of “Learning Scale Free Networks by Reweighted ℓ_1 regularization”

Spectral Dimensionality Reduction via Maximum Entropy

Discussion of “Spectral Dimensionality Reduction via Maximum Entropy”

Discussion of “A conditional game for comparing approximations”

The Discrete Infinite Logistic Normal Distribution for Mixed-Membership Modeling

Discussion of “The Discrete Infinite Logistic Normal Distribution for Mixed-Membership Modeling”