Bandits

1. Implementation of bandit algorithms, with simulation experiment.
2. evaluate E-E policies (Thompson Sampling, UCB, etc) on real data set.

yandex_ranking

Dependency

• Python 3.5
• numpy
• tqdm
• sklearn

Dataset

• Yandex search log link: https://www.kaggle.com/c/yandex-personalized-web-search-challenge#logs-format