用相对似然构建概率集，更准确地表示模型不确定性。
Credal Prediction based on Relative Likelihood
- 基于相对似然阈值筛选可信模型，生成概率分布集合
- 在多个基准数据集上表现优于现有方法，保持预测精度
- 适合需要可靠不确定估计的场景，如医疗或安全系统
以概率分布集合形式进行预测（即可信集），能有效表征学习者的认知不确定性。本文提出一种基于相对似然的理论性可信预测方法：目标是包含所有相对似然超过设定阈值的可相信模型所产生的（条件）概率分布集合。该阈值具有直观意义，可调控预测的正确性与精确性之间的权衡。我们通过改进的集成学习技术来近似此类可信集。在基准数据集上的实验验证了该方法的有效性，其在不牺牲预测性能的前提下，显著提升了不确定性表征能力，并在多个前沿基线方法中表现优异。
原文摘要 · Abstract (English)
Predictions in the form of sets of probability distributions, so-called credal sets, provide a suitable means to represent a learner's epistemic uncertainty. In this paper, we propose a theoretically grounded approach to credal prediction based on the statistical notion of relative likelihood: The target of prediction is the set of all (conditional) probability distributions produced by the collection of plausible models, namely those models whose relative likelihood exceeds a specified threshold. This threshold has an intuitive interpretation and allows for controlling the trade-off between correctness and precision of credal predictions. We tackle the problem of approximating credal sets defined in this way by means of suitably modified ensemble learning techniques. To validate our approach, we illustrate its effectiveness by experiments on benchmark datasets demonstrating superior uncertainty representation without compromising predictive performance. We also compare our method against several state-of-the-art baselines in credal prediction.
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。