给预训练概率电路加个防护罩，让它更抗数据噪声和分布偏移。
PeTeR: Post-Training Robustification of Probabilistic Circuits
- 不重新训练，直接在已有模型上加鲁棒性优化
- 在多个数据集上对随机和对抗扰动都表现更好
- 适合已有的概率电路模型升级，无需新数据
概率电路（PCs）能高效建模复杂联合分布并支持精确推理，但标准似然学习在数据噪声、小样本或分布偏移下易过拟合且泛化脆弱。现有分布鲁棒优化方法仅适用于从头训练，无法用于已有模型。本文提出PeTeR：一种无需数据的后训练鲁棒化框架，可对预训练的PCs进行加固，避免重新训练。在多个密度估计基准上的实验表明，PeTeR能有效提升模型对随机与对抗扰动的鲁棒性，性能媲美甚至超过依赖数据的鲁棒学习基线。
原文摘要 · Abstract (English)
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, small sample sizes, or distribution shifts. This can be mitigated using distributionally-robust optimization which consider worst-case distributions within a Wasserstein ball of the empirical distribution, but current methods are limited to training a model from scratch in this framework. Instead, we propose PeTeR: a novel, data-free post-training framework designed to robustify pre-trained PCs against distribution shifts without retraining from scratch. Empirical evaluations across multiple density estimation benchmarks demonstrate that PeTeR effectively robustifies baseline models against both random and adversarial perturbations, achieving competitive or superior performance to data-dependent robust learning baselines.
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。