Abstract
The analysis of various models in statistical physics relies on the existence of decompositions of measures into mixtures of product-like components, where the goal is to attain a decomposition into measures whose entropy is close to that of the original measure, yet with small correlations between coordinates. We prove a related general result: For every measure \(\mu \) on \({\mathbb {R}}^n\) and every \(\varepsilon > 0\), there exists a decomposition \(\mu = \int \mu _\theta d m(\theta )\) such that \(H(\mu ) - {\mathbb {E}}_{\theta \sim m} H(\mu _\theta ) \le \mathrm {Tr}(\mathrm {Cov}(\mu )) \varepsilon \) and \({\mathbb {E}}_{\theta \sim m} \mathrm {Cov}(\mu _\theta ) \preceq \mathrm {Id}/\varepsilon \). As an application, we derive a general bound for the mean-field approximation of Ising and Potts models, which is in a sense dimension free, in both continuous and discrete settings. In particular, for an Ising model on \(\{\pm \, 1 \}^n\) or on \([-\,1,1]^n\), we show that the deficit between the mean-field approximation and the free energy is at most \(C \frac{1+p}{p} \left( n\Vert J\Vert _{S_p} \right) ^{\frac{p}{1+p}} \) for all \(p>0\), where \(\Vert J\Vert _{S_p}\) denotes the Schatten-p norm of the interaction matrix. For the case \(p=2\), this recovers the result of Jain et al. (Mean-field approximation, convex hierarchies, and the optimality of correlation rounding: a unified perspective. arXiv:1808.07226, 2018), but for an optimal choice of p it often allows to get almost dimension-free bounds.
Similar content being viewed by others
References
Augeri, F.: Nonlinear large deviation bounds with applications to traces of wigner matrices and cycles counts in erdös-renyi graphs (2018). arXiv:1810.01558
Augeri, F.: A transportation approach to the mean-field approximation (2019). arXiv:1903.08021
Austin, T.: The structure of low-complexity gibbs measures on product spaces (2018). arXiv:1810.07278
Bapst, V., Coja-Oghlan, A.: Harnessing the Bethe free energy. Random Struct. Algorithms 49(4), 694–741 (2016)
Basak, A., Mukherjee, S.: Universality of the mean-field for the Potts model. Probab. Theory Relat. Fields 168(3–4), 557–600 (2017)
Chatterjee, S., Dembo, A.: Nonlinear large deviations. Adv. Math. 299, 396–450 (2016)
Coja-Oghlan, A., Krzakala, F., Perkins, W., Zdeborová, L.: Information-theoretic thresholds from the cavity method. Adv. Math. 333, 694–795 (2018)
Coja-Oghlan, A., Perkins, W.: Bethe states of random factor graphs (2017). arXiv:1709.03827
Dembo, A., Montanari, A., Sun, N.: Factor models on locally tree-like graphs. Ann. Probab. 41(6), 4162–4213 (2013)
Eldan, R., Gross, R.: Decomposition of mean-field gibbs distributions into product measures. Electron. J. Probab. (2017). arXiv:1708.05859
Eldan, R.: Thin shell implies spectral gap up to polylog via a stochastic localization scheme. Geom. Funct. Anal. 23(2), 532–569 (2013)
Eldan, R.: Gaussian-width gradient complexity, reverse log-sobolev inequalities and nonlinear large deviations. Geom. Funct. Anal. 28, 1548–1596 (2018)
Eldan, R.: Skorokhod embeddings via stochastic flows on the space of Gaussian measures. Ann. Inst. Henri Poincaré Probab. Stat. 52(3), 1259–1280 (2016)
Eldan, R., Mikulincer, D., Zhai, A.: The central limit theorem in high dimensions: quantitative bounds via martingale embedding (2018). arXiv:1806.09087
Jain, V., Koehler, F., Mossel, E.: Approximating partition functions in constant time (2017). arXiv:1711.01655
Jain, V., Koehler, F., Mossel, E.: The mean-field approximation: information inequalities, algorithms, and complexity. In: Bubeck, S., Perchet, V., Rigollet, P. (eds.) Proceedings of the 31st Conference On Learning Theory, Volume 75 of Proceedings of Machine Learning Research, pp. 1326–1347. PMLR, 06–09 Jul (2018)
Jain, V., Koehler, F., Risteski, A.: Mean-field approximation, convex hierarchies, and the optimality of correlation rounding: a unified perspective (2018). arXiv:1808.07226
Montanari, A.: Estimating random variables from random sparse observations. Eur. Trans. Telecommun. 19, 385–403 (2008)
Manurangsi, P., Raghavendra, P.: A birthday repetition theorem and complexity of approximating dense CSPs. In: 44th International Colloquium on Automata, Languages, and Programming, volume 80 of LIPIcs. Leibniz Int. Proc. Inform., pages Art. No. 78, 15. Schloss Dagstuhl. Leibniz-Zent. Inform., Wadern (2017)
Risteski, A.: How to calculate partition functions using convex programming hierarchies: provable bounds for variational methods. In Feldman, V., Rakhlin, A., Shamir, O. (eds.) 29th Annual Conference on Learning Theory, volume 49 of Proceedings of Machine Learning Research, pp. 1402–1416. Columbia University, New York, New York, USA, 23–26 Jun PMLR (2016)
Raghavendra, P., Tan, N.: Approximating CSPs with global cardinality constraints using SDP hierarchies. In: Proceedings of the Twenty-Third Annual ACM-SIAM Symposium on Discrete Algorithms, pp. 373–384. ACM, New York (2012)
Acknowledgements
I’d like to thank Vishesh Jain, Frederic Koehler and Andrej Risteski for pointing out to me that the approximation in the example of the Curie–Weiss model is sharp, and for several other suggestions regarding the presentation of this manuscript. We are also thankful to Ofer Zeitouni and to the anonymous referee for useful comments and suggestions.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Incumbent of the Elaine Blond Career Development Chair. Supported by a European Research Council Starting Grant (ERC StG) and by the Israel Science Foundation (Grant No. 715/16)
Rights and permissions
About this article
Cite this article
Eldan, R. Taming correlations through entropy-efficient measure decompositions with applications to mean-field approximation. Probab. Theory Relat. Fields 176, 737–755 (2020). https://doi.org/10.1007/s00440-019-00924-2
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00440-019-00924-2