تقرير
Information limits and Thouless-Anderson-Palmer equations for spiked matrix models with structured noise
العنوان: | Information limits and Thouless-Anderson-Palmer equations for spiked matrix models with structured noise |
---|---|
المؤلفون: | Barbier, Jean, Camilli, Francesco, Mondelli, Marco, Xu, Yizhou |
سنة النشر: | 2024 |
المجموعة: | Computer Science Mathematics Condensed Matter Statistics |
مصطلحات موضوعية: | Computer Science - Information Theory, Condensed Matter - Disordered Systems and Neural Networks, Computer Science - Machine Learning, Mathematics - Statistics Theory, 62F15, 82B44 |
الوصف: | We consider a prototypical problem of Bayesian inference for a structured spiked model: a low-rank signal is corrupted by additive noise. While both information-theoretic and algorithmic limits are well understood when the noise is a Gaussian Wigner matrix, the more realistic case of structured noise still proves to be challenging. To capture the structure while maintaining mathematical tractability, a line of work has focused on rotationally invariant noise. However, existing studies either provide sub-optimal algorithms or are limited to special cases of noise ensembles. In this paper, using tools from statistical physics (replica method) and random matrix theory (generalized spherical integrals) we establish the first characterization of the information-theoretic limits for a noise matrix drawn from a general trace ensemble. Remarkably, our analysis unveils the asymptotic equivalence between the rotationally invariant model and a surrogate Gaussian one. Finally, we show how to saturate the predicted statistical limits using an efficient algorithm inspired by the theory of adaptive Thouless-Anderson-Palmer (TAP) equations. |
نوع الوثيقة: | Working Paper |
URL الوصول: | http://arxiv.org/abs/2405.20993 |
رقم الأكسشن: | edsarx.2405.20993 |
قاعدة البيانات: | arXiv |
الوصف غير متاح. |