CADRef: Robust Out-of-Distribution Detection via Class-Aware Decoupled Relative Feature Leveraging
This addresses the safety issue of overconfidence in DNNs for real-world applications, but it is incremental as it builds on existing post-hoc OOD detection methods.
The paper tackles the problem of out-of-distribution (OOD) detection in deep neural networks, which is critical for safe deployment, by proposing two methods (CARef and CADRef) that leverage class-aware relative features, resulting in improvements of up to 3.27% in AUROC and 6.32% in FPR95 over state-of-the-art baselines.
Deep neural networks (DNNs) have been widely criticized for their overconfidence when dealing with out-of-distribution (OOD) samples, highlighting the critical need for effective OOD detection to ensure the safe deployment of DNNs in real-world settings. Existing post-hoc OOD detection methods primarily enhance the discriminative power of logit-based approaches by reshaping sample features, yet they often neglect critical information inherent in the features themselves. In this paper, we propose the Class-Aware Relative Feature-based method (CARef), which utilizes the error between a sample's feature and its class-aware average feature as a discriminative criterion. To further refine this approach, we introduce the Class-Aware Decoupled Relative Feature-based method (CADRef), which decouples sample features based on the alignment of signs between the relative feature and corresponding model weights, enhancing the discriminative capabilities of CARef. Extensive experimental results across multiple datasets and models demonstrate that both proposed methods exhibit effectiveness and robustness in OOD detection compared to state-of-the-art methods. Specifically, our two methods outperform the best baseline by 2.82% and 3.27% in AUROC, with improvements of 4.03% and 6.32% in FPR95, respectively.