Detection of out-of-distribution samples using binary neuron activation patterns

TL;DR

This paper proposes a novel OOD detection method based on binary neuron activation patterns in ReLU networks, demonstrating high performance across multiple architectures and datasets without significant computational overhead.

Contribution

It introduces a new OOD detection technique leveraging binary neuron activation patterns, offering improved accuracy and efficiency over existing confidence-based methods.

Findings

High detection performance across various architectures

Effective on seven image datasets

Low computational overhead

Abstract

Deep neural networks (DNN) have outstanding performance in various applications. Despite numerous efforts of the research community, out-of-distribution (OOD) samples remain a significant limitation of DNN classifiers. The ability to identify previously unseen inputs as novel is crucial in safety-critical applications such as self-driving cars, unmanned aerial vehicles, and robots. Existing approaches to detect OOD samples treat a DNN as a black box and evaluate the confidence score of the output predictions. Unfortunately, this method frequently fails, because DNNs are not trained to reduce their confidence for OOD inputs. In this work, we introduce a novel method for OOD detection. Our method is motivated by theoretical analysis of neuron activation patterns (NAP) in ReLU-based architectures. The proposed method does not introduce a high computational overhead due to the binary representation of the activation patterns extracted from convolutional layers. The extensive empirical evaluation proves its high performance on various DNN architectures and seven image datasets.