A t-distribution based operator for enhancing out of distribution robustness of neural network classifiers

TL;DR

This paper introduces a new t-distribution based operator for neural network classifiers that improves robustness to out-of-distribution samples by better modeling uncertainty, with minimal architectural changes.

Contribution

A novel operator derived from t-distributions is proposed, enhancing out-of-distribution robustness of neural classifiers compared to standard softmax.

Findings

Classifiers with the t-distribution operator outperform softmax-based models on OOD detection.

The new operator provides more reliable uncertainty estimates.

Minimal architectural modifications are required for implementation.

Abstract

Neural Network (NN) classifiers can assign extreme probabilities to samples that have not appeared during training (out-of-distribution samples) resulting in erroneous and unreliable predictions. One of the causes for this unwanted behaviour lies in the use of the standard softmax operator which pushes the posterior probabilities to be either zero or unity hence failing to model uncertainty. The statistical derivation of the softmax operator relies on the assumption that the distributions of the latent variables for a given class are Gaussian with known variance. However, it is possible to use different assumptions in the same derivation and attain from other families of distributions as well. This allows derivation of novel operators with more favourable properties. Here, a novel operator is proposed that is derived using $t$-distributions which are capable of providing a better description of uncertainty. It is shown that classifiers that adopt this novel operator can be more robust to out of distribution samples, often outperforming NNs that use the standard softmax operator. These enhancements can be reached with minimal changes to the NN architecture.