研究人员推出了一种新的训练后框架PeTeR，旨在增强概率电路（PCs）在分布变化下的鲁棒性。与需要从头开始训练的现有方法不同，PeTeR在预训练的PCs上运行，无需额外数据。这种方法旨在缓解由噪声数据或小样本量引起的问题，如过拟合和脆弱的泛化能力。在密度估计基准上的评估表明，PeTeR能有效增强基线模型抵御随机和对抗性扰动，其性能与依赖数据的鲁棒学习技术相当或更优。 AI
影响 在现实世界的嘈杂条件下增强了概率模型的可靠性。
排序理由 该集群描述了一篇详细介绍改进机器学习模型新框架的最新研究论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2607.07671v1 Announce Type: new Abstract: Probabilistic circuits (PCs) can model complex joint distributions while supporting exact and efficient computation of many inference queries. However, standard likelihood-based PC learning is vulnerable to overfitting and fragile g…
Probabilistic circuits (PCs) can model complex joint distributions while supporting exact and efficient computation of many inference queries. However, standard likelihood-based PC learning is vulnerable to overfitting and fragile generalization when confronted with data noise, s…
PeTeR hardens probabilistic circuits without retraining New arXiv preprint PeTeR proposes a data-free way to harden pre-trained probabilistic circuits against distribution shifts without retraining from scratch. https://www. notatechguy.com/peter-hardens- probabilistic-circuits-w…