|
ИСТИНА |
Войти в систему Регистрация |
ИПМех РАН |
||
On one method of non-diagonal regularization in sparse Bayesian learningстатья
Информация о цитировании статьи получена из
Scopus
Дата последнего поиска статьи во внешних источниках: 28 мая 2015 г.
- Авторы: Kropotov D., Vetrov D.
- Сборник: 24th International Conference on Machine Learning, ICML 2007; Corvalis, OR; United States;
- Серия: ACM International Conference Proceeding Series
- Том: 227
- Год издания: 2007
- Место издания: Corvalis, OR
- Первая страница: 457
- Последняя страница: 464
- DOI: 10.1145/1273496.1273554
- Аннотация: In the paper we propose a new type of regularization procedure for training sparse Bayesian methods for classification. Transforming Hessian matrix of log-likelihood function to diagonal form with further regularization of its eigenvectors allows us to optimize evidence explicitly as a product of one-dimensional integrals. The process of automatic regularization coefficients determination then converges in one iteration. We show how to use the proposed approach for Gaussian and Laplace priors. Both algorithms show comparable performance with the state-of-the-art Relevance Vector Machines (RVM) but require less time for training and produce more sparse decision rules (in terms of degrees of freedom).
- Добавил в систему: Кропотов Дмитрий Александрович