提出可微分的不确定性估计框架，让模型自动感知认知不确定性。
Variance-Gated Ensembles: An Epistemic-Aware Framework for Uncertainty Estimation
- 用集成模型统计量构建信噪比门控，动态调节认知不确定性
- 在多个数据集上超越或媲美顶尖信息论方法，计算开销低
- 适合需要可靠置信度评估的工业级模型部署场景
机器学习应用亟需快速可靠的样本级不确定性估计。传统方法通过贝叶斯或近似方法获得预测分布，并将不确定性分解为数据相关（偶然性）与模型相关（认知性）两部分。然而，近期研究指出，当使用有限集成采样和预测分布不匹配时，这种加法分解会失效。本文提出方差门控集成（VGE），一种直观且可微的框架，通过集成统计量计算信噪比门控，注入认知敏感性。VGE提供：(i) 方差门控边缘不确定性（VGMU）评分，将决策边界与集成预测方差耦合；(ii) 方差门控归一化（VGN）层，通过每类可学习的归一化机制，将该机制推广至训练过程。我们推导出闭式向量-雅可比乘积，实现端到端训练中对集成样本均值和方差的梯度传递。VGE在多个基准上达到或超越当前最优的信息论基线，同时保持计算高效。因此，VGE为集成模型提供了实用且可扩展的认知意识不确定性估计方案。
原文摘要 · Abstract (English)
Machine learning applications require fast and reliable per-sample uncertainty estimation. A common approach is to use predictive distributions from Bayesian or approximation methods and additively decompose uncertainty into aleatoric (i.e., data-related) and epistemic (i.e., model-related) components. However, additive decomposition has recently been questioned, with evidence that it breaks down when using finite-ensemble sampling and/or mismatched predictive distributions. This paper introduces Variance-Gated Ensembles (VGE), an intuitive, differentiable framework that injects epistemic sensitivity via a signal-to-noise gate computed from ensemble statistics. VGE provides: (i) a Variance-Gated Margin Uncertainty (VGMU) score that couples decision margins with ensemble predictive variance; and (ii) a Variance-Gated Normalization (VGN) layer that generalizes the variance-gated uncertainty mechanism to training via per-class, learnable normalization of ensemble member probabilities. We derive closed-form vector-Jacobian products enabling end-to-end training through ensemble sample mean and variance. VGE matches or exceeds state-of-the-art information-theoretic baselines while remaining computationally efficient. As a result, VGE provides a practical and scalable approach to epistemic-aware uncertainty estimation in ensemble models.
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。