English(EN) Truncated Out-of-Bag Probability Gain: A Noise-Robust Criterion for Random-Forest Wrapper Feature Selection

新的TOP-Gain方法提高了随机森林特征选择的准确性

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-22 04:00

研究人员开发了一种名为TOP-Gain（截断袋外概率增益）的新方法，用于随机森林中的特征选择。该技术旨在减少袋外预测中的噪声，从而实现更具可复现性和稳定性的特征集选择。在26个数据集的基准测试中，TOP-Gain在ROC-AUC、准确率和Macro-F1方面优于14种基线方法中的13种，同时所需的计算量明显少于同类方法。 AI

影响这种新方法可以提高机器学习模型中特征选择的效率和准确性，从而提高性能并降低计算成本。

排序理由该集群包含一篇研究论文，详细介绍了一种用于机器学习特征选择的新方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 1 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇研究论文，详细介绍了一种用于机器学习特征选择的新方法。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 1 days old; Coverage has settled into its steady-state source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Sejong Oh · 2026-10-22 04:00

Truncated Out-of-Bag Probability Gain: A Noise-Robust Criterion for Random-Forest Wrapper Feature Selection

arXiv:2610.29195v1 Announce Type: new Abstract: Random-Forest wrapper feature selection can use out-of-bag (OOB) predictions as a built-in validation signal, but the criteria that decide each forward step accumulate OOB estimation noise. We propose TOP-Gain (Truncated Out-of-bag …

本文如何被排名

报道来源 [1]

Truncated Out-of-Bag Probability Gain: A Noise-Robust Criterion for Random-Forest Wrapper Feature Selection

相关实体

相关话题