|
ИСТИНА |
Войти в систему Регистрация |
ФНКЦ РР |
||
Breaking Sticks and Ambiguities with Adaptive Skip-gramстатья
- Авторы: Sergey Bartunov, Dmitry Kondrashkin, Anton Osokin, Dmitry Vetrov
- Сборник: Appearing in Proceedings of the 19 th International Conference on Artificial Intelligence and Statistics (AISTATS)
- Том: 51
- Год издания: 2016
- Место издания: JMLR: W&CP Cadiz, Spain
- Первая страница: 130
- Последняя страница: 138
- Аннотация: The recently proposed Skip-gram model is a powerful method for learning high-dimensional word representations that capture rich semantic relationships between words. However, Skip- gram as well as most prior work on learning word representations does not take into account word ambiguity and maintain only a single representa- tion per word. Although a number of Skip-gram modifications were proposed to overcome this limitation and learn multi-prototype word repre- sentations, they either require a known number of word meanings or learn them using greedy heuristic approaches. In this paper we propose the Adaptive Skip-gram model which is a non- parametric Bayesian extension of Skip-gram ca- pable to automatically learn the required num- ber of representations for all words at desired semantic resolution. We derive efficient online variational learning algorithm for the model and empirically demonstrate its efficiency on word- sense induction task.
- Добавил в систему: Журавлев Юрий Иванович