Timezone: »

 
Poster
Leveraging Demonstrations to Improve Online Learning: Quality Matters
Botao Hao · Rahul Jain · Tor Lattimore · Benjamin Van Roy · Zheng Wen

Thu Jul 27 01:30 PM -- 03:00 PM (PDT) @ Exhibit Hall 1 #309
in Poster Session 5 »

We investigate the extent to which offline demonstration data can improve online learning. It is natural to expect some improvement, but the question is how, and by how much? We show that the degree of improvement must depend on the quality of the demonstration data. To generate portable insights, we focus on Thompson sampling (TS) applied to a multi-armed bandit as a prototypical online learning algorithm and model. The demonstration data is generated by an expert with a given competence level, a notion we introduce. We propose an informed TS algorithm that utilizes the demonstration data in a coherent way through Bayes' rule and derive a prior-dependent Bayesian regret bound. This offers insight into how pretraining can greatly improve online performance and how the degree of improvement increases with the expert's competence level. We also develop a practical, approximate informed TS algorithm through Bayesian bootstrapping and show substantial empirical regret reduction through experiments.

[ PDF]  [ Poster] 

Author Information

Botao Hao (Deepmind)
Rahul Jain (USC)
Tor Lattimore (DeepMind)
Benjamin Van Roy (Google)
Zheng Wen (DeepMind)

More from the Same Authors