Abstract
We derive a variational representation for the log-normalizing constant of the posterior distribution in Bayesian linear regression with a uniform spherical prior and an i.i.d. Gaussian design. We work under the “proportional” asymptotic regime, where the number of observations and the number of features grow at a proportional rate. Our representation holds when the variance of the additive noise is sufficiently large, which corresponds to a high-temperature condition in statistical physics. This rigorously establishes the Thouless–Anderson–Palmer (TAP) approximation arising from spin glass theory, and proves a conjecture of (In 2014 IEEE International Symposium on Information Theory (2014) 1499–1503 IEEE) in the special case of the spherical prior (at sufficiently high temperature).
Funding Statement
SS was partially supported by a Harvard Dean’s Competitive Fund Fellowship.
Acknowledgments
SS thanks Sumit Mukherjee for his encouragement during the completion of this manuscript.
Citation
Jiaze Qiu. Subhabrata Sen. "The TAP free energy for high-dimensional linear regression." Ann. Appl. Probab. 33 (4) 2643 - 2680, August 2023. https://doi.org/10.1214/22-AAP1874
Information