让AI在不确定时学会说‘我不确定’，提升安全决策可靠性
Risk-aware Classification via Uncertainty Quantification
- 基于证据深度学习，量化模型预测的不确定性
- 在真实场景中验证，相比现有方法显著降低误判风险
- 适合自动驾驶等高风险领域，帮助系统主动规避危险决策
自主与半自主系统正依赖深度学习优化决策，但深度分类器常对错误预测过度自信，尤其在安全关键领域问题严重。本文提出构建真实世界风险感知分类系统的三大核心要求，并揭示其与证据深度学习（EDL）机制的一致性。通过扩展EDL，使智能体在不确定性与风险并存时具备决策自主权。我们在多个实证场景中严格验证了这些理论创新。相较于现有风险感知分类器，所提方法始终表现更优，凸显其在风险敏感分类策略中的变革潜力。
原文摘要 · Abstract (English)
Autonomous and semi-autonomous systems are using deep learning models to improve decision-making. However, deep classifiers can be overly confident in their incorrect predictions, a major issue especially in safety-critical domains. The present study introduces three foundational desiderata for developing real-world risk-aware classification systems. Expanding upon the previously proposed Evidential Deep Learning (EDL), we demonstrate the unity between these principles and EDL's operational attributes. We then augment EDL empowering autonomous agents to exercise discretion during structured decision-making when uncertainty and risks are inherent. We rigorously examine empirical scenarios to substantiate these theoretical innovations. In contrast to existing risk-aware classifiers, our proposed methodologies consistently exhibit superior performance, underscoring their transformative potential in risk-conscious classification strategies.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。