MINE: Mutual Information Neural Estimation

Author(s): Sherwin Chen

Originally published on Towards AI.

Estimating mutual information using arbitrary neural networks through MINE

Source: istock.com/ipopba

Mutual information, also known as information gain, has been successfully used in the context of deep learning(which we will see soon) and deep reinforcement learning(e.g., VIME, EMI) to measure/enhance the coupling between two representations. In this article, we discuss in detail a neural estimator named MINE(Mutual Information Neural Estimation), published by Mohamed Ishmael Belghazi et al. in ICML 2018, that allows us to directly estimate the mutual information.

This article is comprised of three parts: We first introduce the concept of mutual information, building some intuition for a better understanding of what we are dealing with. Then we present the… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI