arXiv cs.AI· Alberto Foresti, Ivan Butakov, Alexander Tolmachev, Giulio Franzese, Alexey Frolov, Pietro Michiardi·· 5 小时前AI 评分26
面向互信息估计的多样化综合基准研究
Towards Diverse and Comprehensive Benchmarks for Mutual Information Estimation
AI 导读
一项 arXiv 研究提出基于统一 copula 理论视角的互信息估计基准框架，包含 copula-first 与 marginals-first 两类测试，后者将真实图像数据与受控依赖结构结合。研究系统评估了非参数、变分和生成三类估计器，发现没有通用最优者：每类估计器都会在特定设置下系统性优于其他方法。作者据此识别出根本性估计障碍并提出了更能施压这些局限的新测试，代码已开源。
正文 · 原文
Abstract:Mutual information estimation is a central problem in machine learning and statistics; however, existing benchmarks typically evaluate estimators on simplified, low-dimensional distributions, leaving their performance on complex, realistic data largely unexplored. We address this gap with a comprehensive benchmarking framework grounded in a unified copula-theoretic perspective that subsumes existing benchmarks as special cases. Within this framework, we propose two complementary families of tests: a copula-first family that systematically varies ground-truth mutual information, dimensionality, and marginal complexity using synthetic and flow-based transformations; and a marginals-first family that couples real-world image data with controlled dependency structures, extending the classic same-class-pairing paradigm. We use this suite to extensively evaluate three classes of estimators: non-parametric, variational, and generative. Contrary to prevailing assumptions, our results indicate that there is no universal winner: each category can systematically outperform all other estimators under specific setups. By analyzing these cases, we identify fundamental estimation barriers and propose new tests that more effectively stress these specific limitations. We share the open source code at this https URL.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Information Theory (cs.IT) |
| Cite as: | arXiv:2607.03487 [cs.LG] |
| (or arXiv:2607.03487v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2607.03487 arXiv-issued DOI via DataCite |
Submission history
From: Alberto Foresti [view email]
[v1]
Fri, 3 Jul 2026 16:55:55 UTC (162 KB)
[v2]
Fri, 30 Oct 2026 08:36:14 UTC (266 KB)
来源：arXiv cs.AI · arxiv.org