PulseAugur
中
实时 08:17:50
English(EN) What Survival Benchmarks Don't Tell You: Impact of Model Selection and Dataset Regimes

生存基准测试的局限性暴露:深度学习、数据集机制和评估指标分析

一篇新发表在arXiv上的研究论文探讨了当前生存基准测试方法的局限性,特别是它们对一致性指数(C-index)的依赖以及对深度学习(DL)方法的代表性不足。该研究进行了大规模基准测试,比较了10种不同方法在49个数据集上的表现,并将模型选择标准(C-index、集成Brier分数(IBS)或平均绝对误差(MAE-PO))视为一个变量。研究结果表明,与C-index相比,优化IBS能够显著提高复杂DL和基于树的模型在概率准确性和校准方面的表现。研究还强调,数据集特征从根本上决定了模型性能,DL模型在样本量增加时表现良好,但在高维特征空间中会遇到困难,并指出重度删失可能人为地夸大校准指标的漏洞。 AI

影响 挑战了当前生存分析的评估标准,主张更透明地报告方法论,并可能影响未来深度学习模型的基准设计。

排序理由 该集群包含一篇详细介绍生存分析方法的大规模基准测试和分析的研究论文。[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 →

生存基准测试的局限性暴露:深度学习、数据集机制和评估指标分析

本文如何被排名

Signal score
23 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群包含一篇详细介绍生存分析方法的大规模基准测试和分析的研究论文。[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
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.LG TIER_1 English(EN) · Sandra Dening, Emre Kavak, Christian Wachinger, George H. Chen ·

    生存基准测试未告知你的信息:模型选择和数据集制度的影响

    arXiv:2611.24837v1 Announce Type: new Abstract: Despite the proliferation of survival analysis methods, current benchmarks often operate under narrow conditions: they heavily rely on the concordance index (C-index) for evaluation and model selection, and frequently either exclude…