# MEMe: An Accurate Maximum Entropy Method for Efficient Approximations in Large-Scale Machine Learning

## TL;DR

This paper introduces MEMe, a robust maximum entropy method that efficiently approximates large-scale machine learning problems by handling numerous moments, demonstrating superiority in log determinant estimation and Bayesian optimization.

## Contribution

The paper presents a novel maximum entropy algorithm capable of efficiently managing hundreds of moments, with demonstrated advantages over existing methods.

## Key findings

- Superior accuracy in log determinant estimation
- Effective in information-theoretic Bayesian optimization
- Handles hundreds of moments efficiently

## Abstract

Efficient approximation lies at the heart of large-scale machine learning problems. In this paper, we propose a novel, robust maximum entropy algorithm, which is capable of dealing with hundreds of moments and allows for computationally efficient approximations. We showcase the usefulness of the proposed method, its equivalence to constrained Bayesian variational inference and demonstrate its superiority over existing approaches in two applications, namely, fast log determinant estimation and information-theoretic Bayesian optimisation.