MINE: Mutual Information Neural Estimation
This work addresses a fundamental challenge in machine learning for researchers and practitioners by providing a flexible and efficient method for mutual information estimation, with incremental improvements in specific applications.
The authors tackled the problem of estimating mutual information between high-dimensional continuous variables by proposing a neural estimator (MINE) that is scalable and trainable via back-propagation, achieving strong consistency and demonstrating improved performance in applications like generative models and supervised classification.
We argue that the estimation of mutual information between high dimensional continuous random variables can be achieved by gradient descent over neural networks. We present a Mutual Information Neural Estimator (MINE) that is linearly scalable in dimensionality as well as in sample size, trainable through back-prop, and strongly consistent. We present a handful of applications on which MINE can be used to minimize or maximize mutual information. We apply MINE to improve adversarially trained generative models. We also use MINE to implement Information Bottleneck, applying it to supervised classification; our results demonstrate substantial improvement in flexibility and performance in these settings.