Empirical Evidences for the Effects of Feature Diversity in Open Set Recognition and Continual Learning

TL;DR

This paper provides empirical evidence that increasing feature diversity enhances open set recognition and continual learning by improving detection of novel classes and retention of learned data.

Contribution

It is the first to empirically demonstrate the positive effects of feature diversity on open set recognition and continual learning performance.

Findings

Enhanced feature diversity improves open set sample recognition.

Increased feature diversity aids in retaining previous knowledge.

Feature diversity facilitates integration of new data in continual learning.

Abstract

Open set recognition (OSR) and continual learning are two critical challenges in machine learning, focusing respectively on detecting novel classes at inference time and updating models to incorporate the new classes. While many recent approaches have addressed these problems, particularly OSR, by heuristically promoting feature diversity, few studies have directly examined the role that feature diversity plays in tackling them. In this work, we provide empirical evidence that enhancing feature diversity improves the recognition of open set samples. Moreover, increased feature diversity also facilitates both the retention of previously learned data and the integration of new data in continual learning. We hope our findings can inspire further research into both practical methods and theoretical understanding in these domains.