Effects of Common Regularization Techniques on Open-Set Recognition

TL;DR

This paper investigates how common regularization techniques influence the performance of neural networks in open-set recognition tasks, highlighting that regularization can significantly improve the ability to detect unknown inputs across various datasets.

Contribution

It provides empirical evidence that regularization methods enhance open-set recognition performance and offers new insights into the accuracy-open-set performance relationship.

Findings

Regularization improves open-set detection accuracy.

Regularization techniques enhance model generalization to unknown classes.

Insights into the trade-off between accuracy and open-set detection performance.

Abstract

In recent years there has been increasing interest in the field of Open-Set Recognition, which allows a classification model to identify inputs as "unknown" when it encounters an object or class not in the training set. This ability to flag unknown inputs is of vital importance to many real world classification applications. As almost all modern training methods for neural networks use extensive amounts of regularization for generalization, it is therefore important to examine how regularization techniques impact the ability of a model to perform Open-Set Recognition. In this work, we examine the relationship between common regularization techniques and Open-Set Recognition performance. Our experiments are agnostic to the specific open-set detection algorithm and examine the effects across a wide range of datasets. We show empirically that regularization methods can provide significant improvements to Open-Set Recognition performance, and we provide new insights into the relationship between accuracy and Open-Set performance.