English

Sparsity-Inducing Categorical Prior Improves Robustness of the Information Bottleneck

Machine Learning 2022-10-31 v2 Machine Learning Methodology

Abstract

The information bottleneck framework provides a systematic approach to learning representations that compress nuisance information in the input and extract semantically meaningful information about predictions. However, the choice of a prior distribution that fixes the dimensionality across all the data can restrict the flexibility of this approach for learning robust representations. We present a novel sparsity-inducing spike-slab categorical prior that uses sparsity as a mechanism to provide the flexibility that allows each data point to learn its own dimension distribution. In addition, it provides a mechanism for learning a joint distribution of the latent variable and the sparsity and hence can account for the complete uncertainty in the latent space. Through a series of experiments using in-distribution and out-of-distribution learning scenarios on the MNIST, CIFAR-10, and ImageNet data, we show that the proposed approach improves accuracy and robustness compared to traditional fixed-dimensional priors, as well as other sparsity induction mechanisms for latent variable models proposed in the literature.

Keywords

Cite

@article{arxiv.2203.02592,
  title  = {Sparsity-Inducing Categorical Prior Improves Robustness of the Information Bottleneck},
  author = {Anirban Samaddar and Sandeep Madireddy and Prasanna Balaprakash and Tapabrata Maiti and Gustavo de los Campos and Ian Fischer},
  journal= {arXiv preprint arXiv:2203.02592},
  year   = {2022}
}