Shaping Parameter Contribution Patterns for Out-of-Distribution Detection

2026-03-07

AI Summary

This work addresses the overconfidence of deep neural networks in out-of-distribution (OOD) detection, which often stems from reliance on sparse patterns of parameter contributions. To mitigate this issue, the authors propose SPCP, a method that regularizes parameter contributions during standard classification training via a dynamic thresholding mechanism, thereby encouraging the model to learn dense contribution patterns aligned with decision boundaries. Notably, SPCP requires neither additional network components nor post-hoc processing, and it represents the first approach to explicitly shape parameter contribution patterns to enhance OOD robustness. Experimental results demonstrate that SPCP significantly reduces confidence on OOD samples across various detection settings while preserving strong in-distribution classification performance.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Out-of-distribution (OOD) detection is a well-known challenge due to deep models often producing overconfident. In this paper, we reveal a key insight that trained classifiers tend to rely on sparse parameter contribution patterns, meaning that only a few dominant parameters drive predictions. This brittleness can be exploited by OOD inputs that anomalously trigger these parameters, resulting in overconfident predictions. To address this issue, we propose a simple yet effective method called Shaping Parameter Contribution Patterns (SPCP), which enhances OOD detection robustness by encouraging the classifier to learn boundary-oriented dense contribution patterns. Specifically, SPCP operates during training by rectifying excessively high parameter contributions based on a dynamically estimated threshold. This mechanism promotes the classifier to rely on a broader set of parameters for decision-making, thereby reducing the risk of overconfident predictions caused by anomalously triggered parameters, while preserving in-distribution (ID) performance. Extensive experiments under various OOD detection setups verify the effectiveness of SPCP.

Problem

• out-of-distribution detection
• overconfident prediction
• parameter contribution
• deep models
• distribution shift

Innovation

• out-of-distribution detection
• parameter contribution patterns
• dense contribution
• overconfidence mitigation
• SPCP