资源分类
》 资源 》论文》A KL-LUCB Bandit Algorithm for Large-Scale Crowdsourcing

A KL-LUCB Bandit Algorithm for Large-Scale Crowdsourcing

2020-02-10 | |  479 |   118 |   0

Abstract 

This paper focuses on best-arm identification in multi-armed bandits with bounded rewards. We develop an algorithm that is a fusion of lil-UCB and KL-LUCB, offering the best qualities of the two algorithms in one method. This is achieved by proving a novel anytime confidence bound for the mean of bounded distributions, which is the analogue of the LIL-type bounds recently developed for sub-Gaussian distributions. We corroborate our theoretical results with numerical experiments based on the New Yorker Cartoon Caption Contest. 

上一篇：Flexible statistical inference for mechanistic models of neural dynamics

下一篇：A Decomposition of Forecast Error in Prediction Markets

用户评价
全部评价

热门资源

  • Learning to Predi...

    Much of model-based reinforcement learning invo...

  • dynamical system ...

    allows to preform manipulations of heavy or bul...

  • Rating-Boosted La...

    The performance of a recommendation system reli...

  • The Variational S...

    Unlike traditional images which do not offer in...

  • Learning to learn...

    The move from hand-designed features to learned...