Semantic Driven Energy based Out-of-Distribution Detection

TL;DR

This paper introduces a semantic-driven energy-based method for out-of-distribution detection that improves accuracy by learning class representations and a novel loss function, achieving state-of-the-art results on benchmarks.

Contribution

The paper proposes a new end-to-end trainable energy-based OOD detection method with a novel Cluster Focal Loss for better class representation learning.

Findings

Significantly reduces false positive rates on CIFAR benchmarks.

Achieves state-of-the-art OOD detection performance.

Extends framework to object detection with improved results.

Abstract

Detecting Out-of-Distribution (OOD) samples in real world visual applications like classification or object detection has become a necessary precondition in today's deployment of Deep Learning systems. Many techniques have been proposed, of which Energy based OOD methods have proved to be promising and achieved impressive performance. We propose semantic driven energy based method, which is an end-to-end trainable system and easy to optimize. We distinguish in-distribution samples from out-distribution samples with an energy score coupled with a representation score. We achieve it by minimizing the energy for in-distribution samples and simultaneously learn respective class representations that are closer and maximizing energy for out-distribution samples and pushing their representation further out from known class representation. Moreover, we propose a novel loss function which we call Cluster Focal Loss(CFL) that proved to be simple yet very effective in learning better class wise cluster center representations. We find that, our novel approach enhances outlier detection and achieve state-of-the-art as an energy-based model on common benchmarks. On CIFAR-10 and CIFAR-100 trained WideResNet, our model significantly reduces the relative average False Positive Rate(at True Positive Rate of 95%) by 67.2% and 57.4% respectively, compared to the existing energy based approaches. Further, we extend our framework for object detection and achieve improved performance.