Neural Mutual Information Estimation with Vector Copulas
Not on the map. Why?
Abstract (arXiv)
Estimating mutual information (MI) is a fundamental task in data science and machine learning. Existing estimators mainly rely on either highly flexible models (e.g., neural networks), which require large amounts of data, or overly simplified models (e.g., Gaussian copula), which fail to capture complex distributions. Drawing upon recent vector copula theory, we propose a principled interpolation between these two extremes to achieve a better trade-off between complexity and capacity. Experiments on state-of-the-art synthetic benchmarks and real-world data with diverse modalities demonstrate the advantages of the proposed estimator.
Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 44 of its 73 references that are in the map. First posted on arXiv 23 Oct 2025.
Builds on
44 of 73 references.
References · 44
- 2024The Llama 3 Herd of Modelscited by 14,715
- 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
- 2016Densely Connected Convolutional Networkscited by 6,720
- 2022Flow Matching for Generative Modelingcited by 5,702
- 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,323
- 2018Learning deep representations by mutual information estimation and maximizationcited by 1,273
- 2016Deep Variational Information Bottleneckcited by 930
- 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 860
- 2017Masked Autoregressive Flow for Density Estimationcited by 684
- 2018MINE: Mutual Information Neural Estimationcited by 683
- 2019Neural Spline Flowscited by 596
- 2016Error bounds for approximations with deep ReLU networkscited by 574
32 more
- 2019On Variational Bounds of Mutual Informationcited by 559
- 2019On Mutual Information Maximization for Representation Learningcited by 316
- 2020CLUB: A Contrastive Log-ratio Upper Bound of Mutual Informationcited by 235
- 2018Formal Limitations on the Measurement of Mutual Informationcited by 189
- 2019Understanding the Limitations of Variational Mutual Information Estimatorscited by 140
- 2020Telescoping Density-Ratio Estimationcited by 108
- 2020Neural Approximate Sufficient Statistics for Implicit Modelscited by 74
- 2020On Mutual Information in Contrastive Learning for Visual Representationscited by 54
- 2013Equitability, mutual information, and the maximal information coefficientcited by 45
- 2019CCMI : Classifier based Conditional Mutual Information Estimationcited by 37
- 2023Beyond Normal: On the Evaluation of Mutual Information Estimatorscited by 37
- 2023MINDE: Mutual Information Neural Diffusion Estimationcited by 31
- 2023TACTiS-2: Better, Faster, Simpler Attentional Copulas for Multivariate Time Seriescited by 26
- 2021Unsupervised Hashing with Contrastive Information Bottleneckcited by 24
- 2024Mutual Information Estimation via Normalizing Flowscited by 22
- 2024Approximating mutual information of high-dimensional variables using learned representationscited by 21
- 2023Information Bottleneck Analysis of Deep Neural Networks via Lossy Compressioncited by 19
- 2021Tight Mutual Information Estimation With Contrastive Fenchel-Legendre Optimizationcited by 18
- 2023Estimating the Density Ratio between Distributions with High Discrepancy using Multinomial Logistic Regressioncited by 17
- 2023Mutual Information Estimation via f-Divergence and Data Derangementscited by 17
- 2022Copula Density Neural Estimationcited by 13
- 2024Inf2Guard: An Information-Theoretic Framework for Learning Privacy-Preserving Representations against Inference Attackscited by 13
- 2022Diffeomorphic Information Neural Estimationcited by 10
- 2023Scalable Infomin Learningcited by 8
- 2024Your copula is a classifier in disguise: classification-based copula density estimationcited by 8
- 2018Information estimation using nonparametric copulascited by 7
- 2019Adaptive Gaussian Copula ABCcited by 7
- 2021MIND: Inductive Mutual Information Estimation, A Convex Maximum-Entropy Copula Approachcited by 6
- 2020DEMI: Discriminative Estimator of Mutual Informationcited by 5
- 2017Nonparanormal Information Estimationcited by 4
- 2023Towards a universal representation of statistical dependencecited by 4
- 2022fastMI: a fast and consistent copula-based estimator of mutual informationcited by 3
Built on by
Citing papers per year: 2025: 2; 2026: 2.
4 papers.
- 2025FMMI: Flow Matching Mutual Information Estimationcited by 4
- 2025MIST: Mutual Information Estimation Via Supervised Trainingcited by 1
- 2026InfoAtlas: A Foundation Model for Zero-Shot Statistical Dependence Estimatecited by 1
- 2026Towards Diverse and Comprehensive Benchmarks for Mutual Information Estimationcited by 0
- 2026Towards Diverse and Comprehensive Benchmarks for Mutual Information Estimation3 Jul 2026
- 2026InfoAtlas: A Foundation Model for Zero-Shot Statistical Dependence Estimate29 May 2026
- 2025MIST: Mutual Information Estimation Via Supervised Training24 Nov 2025
- 2025FMMI: Flow Matching Mutual Information Estimation11 Nov 2025