Impact of Representation Learning in Linear Bandits

Jiaqi Yang

Wei Hu (胡威)

[0]

Jason D. Lee

[0]

Simon Shaolei Du (杜少雷)

[0]

international conference on learning representations, 2020.

Cited by: 0|Views38

Keywords:

sequential decisionnew algorithmdkn tmulti taskconfidence intervalMore(9+)

Weibo:

We show representation learning provably improves multi-task linear bandits.

Abstract:

We study how representation learning can improve the efficiency of bandit problems. We study the setting where we play T linear bandits with dimension d concurrently, and these T bandit tasks share a common k(≪d) dimensional linear representation. For the finite-action setting, we present a new algorithm which achieves O~(TkN+dkNT) regret...More

Code:

Data:

Full Text

View via Publisher