Learning capability of the rescaled pure greedy algorithm with non-iid sampling

Autor: Qin Guo, Binlei Cai
Jazyk: angličtina
Rok vydání: 2023
Předmět:
Zdroj: Electronic Research Archive, Vol 31, Iss 3, Pp 1387-1404 (2023)
Druh dokumentu: article
ISSN: 2688-1594
DOI: 10.3934/era.2023071?viewType=HTML
Popis: We consider the rescaled pure greedy learning algorithm (RPGLA) with the dependent samples drawn according to a non-identical sequence of probability distributions. The generalization performance is provided by applying the independent-blocks technique and adding the drift error. We derive the satisfactory learning rate for the algorithm under the assumption that the process satisfies stationary $ \beta $-mixing, and also find that the optimal rate $ O(n^{-1}) $ can be obtained for i.i.d. processes.
Databáze: Directory of Open Access Journals
Nepřihlášeným uživatelům se plný text nezobrazuje