Neural_Dueling_Bandits

My Preliminary

In 多臂老虎机 - 动手学强化学习, we get to know about the problem definition of multi-armed bandit, some conception, $\varepsilon$-greedy algorithm, UCB and Thompson sampling.

When it comes to Thompson sampling, check these two courses: The Beta distribution in 12 minutes! and Thompson sampling, one armed bandits, and the Beta distribution. You can search them on Bilibili in homeland.

After watching two videos metioned above, go back to the Thompson sampling in 多臂老虎机 - 动手学强化学习

And then read “A Contextual-Bandit Approach to Personalized News Article Recommendation” to get to know about LinUCB.

Introduction

• identifying the most appropriate web-based content at the best time for individual users

Methodology