DeepAI AI Chat
Log In Sign Up

Neuron Activation Coverage: Rethinking Out-of-distribution Detection and Generalization

by   Yibing Liu, et al.
City University of Hong Kong
Association for Computing Machinery

The out-of-distribution (OOD) problem generally arises when neural networks encounter data that significantly deviates from the training data distribution, , in-distribution (InD). In this paper, we study the OOD problem from a neuron activation view. We first formulate neuron activation states by considering both the neuron output and its influence on model decisions. Then, we propose the concept of neuron activation coverage (NAC), which characterizes the neuron behaviors under InD and OOD data. Leveraging our NAC, we show that 1) InD and OOD inputs can be naturally separated based on the neuron behavior, which significantly eases the OOD detection problem and achieves a record-breaking performance of 0.03 outperforming the previous best method by 20.67 between NAC and model generalization ability consistently holds across architectures and datasets, which enables a NAC-based criterion for evaluating model robustness. By comparison with the traditional validation criterion, we show that NAC-based criterion not only can select more robust models, but also has a stronger correlation with OOD test performance.


page 1

page 2

page 3

page 4


IC Neuron: An Efficient Unit to Construct Neural Networks

As a popular machine learning method, neural networks can be used to sol...

Generalization Beyond Feature Alignment: Concept Activation-Guided Contrastive Learning

Learning invariant representations via contrastive learning has seen sta...

The Compact Support Neural Network

Neural networks are popular and useful in many fields, but they have the...

You Can't See the Forest for Its Trees: Assessing Deep Neural Network Testing via NeuraL Coverage

This paper summarizes eight design requirements for DNN testing criteria...

Neuron with Steady Response Leads to Better Generalization

Regularization can mitigate the generalization gap between training and ...

Runtime Monitoring Neuron Activation Patterns

For using neural networks in safety critical domains, it is important to...

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

Deep neural networks (DNN) have outstanding performance in various appli...