在粒子物理数据中首次验证过参数化模型的双下降现象
Double Descent and Overparameterization in Particle Physics Data
- 用过参数化模型测试双下降现象在粒子物理中的表现
- 发现模型容量超过插值阈值后泛化误差反而下降
- 为高能物理数据分析提供新思路，适合机器学习研究者
近期机器学习任务中观察到：过度参数化的模型——其容量足以轻松跨越插值阈值——相比经典偏差-方差权衡区域，反而表现出更优的泛化性能。本文首次在粒子物理数据中验证了这一现象，系统探索了双下降出现的条件以及过参数化在何种情况下带来性能提升。
原文摘要 · Abstract (English)
Recently, the benefit of heavily overparameterized models has been observed in machine learning tasks: models with enough capacity to easily cross the \emph{interpolation threshold} improve in generalization error compared to the classical bias-variance tradeoff regime. We demonstrate this behavior for the first time in particle physics data and explore when and where `double descent' appears and under which circumstances overparameterization results in a performance gain.
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。