Semantic Novelty Detection via Relational Reasoning

TL;DR

This paper introduces a relational reasoning-based representation learning method for semantic novelty detection that effectively identifies unknown categories without requiring model fine-tuning, suitable for resource-constrained and privacy-sensitive applications.

Contribution

It proposes a novel relational reasoning paradigm for semantic novelty detection that enhances open-set recognition by focusing on measuring semantic similarity, enabling plug-and-play integration.

Findings

Outperforms state-of-the-art methods in novelty detection tasks.

Enables conversion of closed-set models into open-set models.

Effective in privacy-sensitive and resource-constrained scenarios.

Abstract

Semantic novelty detection aims at discovering unknown categories in the test data. This task is particularly relevant in safety-critical applications, such as autonomous driving or healthcare, where it is crucial to recognize unknown objects at deployment time and issue a warning to the user accordingly. Despite the impressive advancements of deep learning research, existing models still need a finetuning stage on the known categories in order to recognize the unknown ones. This could be prohibitive when privacy rules limit data access, or in case of strict memory and computational constraints (e.g. edge computing). We claim that a tailored representation learning strategy may be the right solution for effective and efficient semantic novelty detection. Besides extensively testing state-of-the-art approaches for this task, we propose a novel representation learning paradigm based on relational reasoning. It focuses on learning how to measure semantic similarity rather than recognizing known categories. Our experiments show that this knowledge is directly transferable to a wide range of scenarios, and it can be exploited as a plug-and-play module to convert closed-set recognition models into reliable open-set ones.