Can I read UCB and Thompson Sampling Algorithms on EtoBox?

UCB and Thompson Sampling Algorithms by SAURABH VISHWAKARMA is a document available to read on EtoBox.

What is UCB and Thompson Sampling Algorithms about?

The document discusses the UCB algorithm for multi-armed bandit problems. It provides the formal description of the UCB algorithm and derives instance-dependent and instance-independent regret bounds for UCB, showing they match the bounds for successive elimination. It also briefly introduces Thompson sampling, which maintains probability distributions over the arm means rather than point estimates.

Author

SAURABH VISHWAKARMA

Language

EN