新的贝叶斯损失函数可识别机器学习模型中的数据污染

作者 PulseAugur 编辑部 · [2 个来源] · 2026-06-15 10:25

研究人员开发了神经贝叶斯异常缓解（NBAM），这是一种新颖的损失函数，旨在提高监督机器学习模型对数据污染的鲁棒性。NBAM 不仅使模型能够容忍损坏的数据（类似于 Huber 或 Student's t-test 等现有鲁棒损失），还能作为无监督分类器来识别哪些特定观测值已被损坏。该方法利用贝叶斯潜在开关混合模型来实现这一目标，在具有显著污染率的 CIFAR-10 数据集上表现优于基线鲁棒损失。 AI

影响这项研究介绍了一种提高机器学习数据质量的方法，有望在真实世界中通常嘈杂的数据集上训练出更可靠的模型。

排序理由该集群包含一篇学术论文，详细介绍了新的研究方法及其在基准数据集上的评估。

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.LG TIER_1 English(EN) · S. A. K. Leeney, W. J. Handley, H. T. J. Bevins, E. de Lera Acedo · 2026-06-16 04:00

Neural Bayesian Anomaly Mitigation: A Robust Loss that Doubles as an Unsupervised Contamination Classifier

arXiv:2606.16524v1 Announce Type: new Abstract: Engineered robust losses such as Huber, Student-$t$, and generalised cross-entropy make supervised models tolerant of contamination but cannot answer which observations are corrupted. We introduce Neural Bayesian Anomaly Mitigation …
arXiv stat.ML TIER_1 English(EN) · E. de Lera Acedo · 2026-06-15 10:25

Neural Bayesian Anomaly Mitigation: A Robust Loss that Doubles as an Unsupervised Contamination Classifier

Engineered robust losses such as Huber, Student-$t$, and generalised cross-entropy make supervised models tolerant of contamination but cannot answer which observations are corrupted. We introduce Neural Bayesian Anomaly Mitigation (NBAM), a general-purpose drop-in loss derived f…

报道来源 [2]

Neural Bayesian Anomaly Mitigation: A Robust Loss that Doubles as an Unsupervised Contamination Classifier

Neural Bayesian Anomaly Mitigation: A Robust Loss that Doubles as an Unsupervised Contamination Classifier

相关实体

相关话题