English

Learning deep representations by mutual information estimation and maximization

Machine Learning 2019-02-25 v5 Machine Learning

Abstract

In this work, we perform unsupervised learning of representations by maximizing mutual information between an input and the output of a deep neural network encoder. Importantly, we show that structure matters: incorporating knowledge about locality of the input to the objective can greatly influence a representation's suitability for downstream tasks. We further control characteristics of the representation by matching to a prior distribution adversarially. Our method, which we call Deep InfoMax (DIM), outperforms a number of popular unsupervised learning methods and competes with fully-supervised learning on several classification tasks. DIM opens new avenues for unsupervised learning of representations and is an important step towards flexible formulations of representation-learning objectives for specific end-goals.

Keywords

Cite

@article{arxiv.1808.06670,
  title  = {Learning deep representations by mutual information estimation and maximization},
  author = {R Devon Hjelm and Alex Fedorov and Samuel Lavoie-Marchildon and Karan Grewal and Phil Bachman and Adam Trischler and Yoshua Bengio},
  journal= {arXiv preprint arXiv:1808.06670},
  year   = {2019}
}

Comments

Accepted as an oral presentation at the International Conference for Learning Representations (ICLR), 2019

R2 v1 2026-06-23T03:38:53.980Z