Efficient Sample-optimal Learning of Gaussian Tree Models via Sample-optimal Testing of Gaussian Mutual Information
Preprint
Abstract
Learning high-dimensional distributions is a significant challenge in machine learning and statistics. Classical research has mostly concentrated on asymptotic analysis of such data under suitable assumptions. While existing works [Bhattacharyya et al.: SICOMP 2023, Daskalakis et al.: STOC 2021, Choo et al.: ALT 2024] focus on discrete distributions, the current work addresses the tree structure learning problem for Gaussian distributions, providing efficient algorithms with solid theoretical guarantees. This is crucial as real-world distributions are often continuous and differ from the discrete scenarios studied in prior works. In this work, we design a conditional mutual information tester for Gaussian random variables that can test whether two Gaussian random variables are independent, or their conditional mutual information is at least $\varepsilon$, for some parameter $\varepsilon \in (0,1)$ using $\mathcal{O}(\varepsilon^{-1})$ samples which we show to be near-optimal. In contrast, an additive estimation would require $Ω(\varepsilon^{-2})$ samples. Our upper bound technique uses linear regression on a pair of suitably transformed random variables. Importantly, we show that the chain rule of conditional mutual information continues to hold for the estimated (conditional) mutual information. As an application of such a mutual information tester, we give an efficient $\varepsilon$-approximate structure-learning algorithm for an $n$-variate Gaussian tree model that takes $\widetildeΘ(n\varepsilon^{-1})$ samples which we again show to be near-optimal. In contrast, when the underlying Gaussian model is not known to be tree-structured, we show that $\widetilde{Θ}(n^2\varepsilon^{-2})$ samples are necessary and sufficient to output an $\varepsilon$-approximate tree structure. We perform extensive experiments that corroborate our theoretical convergence bounds.
Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Conditional independence testing for discrete distributions: Beyond χ2- and G-testsElectronic Journal of Statistics, 2024
- Local permutation tests for conditional independenceThe Annals of Statistics, 2022
- Minimax optimal conditional independence testingThe Annals of Statistics, 2021
- Learning a tree-structured Ising model in order to make predictionsThe Annals of Statistics, 2020
- Taming big probability distributionsXRDS: Crossroads, The ACM Magazine for Students, 2012
- A Constrainedℓ1Minimization Approach to Sparse Precision Matrix EstimationJournal of the American Statistical Association, 2011
- Sparse inverse covariance estimation with the graphical lassoBiostatistics, 2007
- Maximum likelihood bounded tree-width Markov networksArtificial Intelligence, 2003
- Approximating discrete probability distributions with dependence treesIEEE Transactions on Information Theory, 1968