×
Feb 12, 2024 · In this work, we prove high-probability generalization bounds for heavy-tailed SDEs which do not contain any nontrivial information theoretic terms.
To achieve this goal, we develop new proof techniques based on estimating the entropy flows associated with the so-called fractional Fokker-Planck equation (a ...
Jun 3, 2024 · This paper addresses the important problem of understanding the generalization properties of heavy-tailed stochastic optimization algorithms, ...
People also ask
Generalization Bounds for Heavy-Tailed SDEs through the Fractional Fokker-Planck Equation · Benjamin Dupuis, Umut Simsekli. Published: 01 May 2024, Last ...
In statistical learning theory, a generalization bound usually involves a complexity measure imposed by the considered theoretical framework. 0. 19 Feb 2024.
Generalization Bounds for Heavy-Tailed SDEs through the Fractional Fokker-Planck Equation. B Dupuis, U Şimşekli. arXiv preprint arXiv:2402.07723, 2024. 2024.
In statistical learning theory, a generalization bound usually involves a complexity measure imposed by the considered theoretical framework. 0. 19 Feb 2024.
Addressing these drawbacks, in this work, we prove high-probability generalization bounds for heavy-tailed SDEs which do not contain any nontrivial information ...
Simsekli, "Generalization Bounds for Heavy-Tailed SDEs through the Fractional Fokker-Planck Equation", arXiv, 2024. F. Schaipp, G. Garrigos, U. Simsekli ...
Addressing these drawbacks, in this work, we prove high-probability generalization bounds for heavy-tailed SDEs which do not contain any nontrivial information ...