The Impact of Batch Learning in Stochastic Bandits
arXiv:2111.02071
Abstract
We consider a special case of bandit problems, namely batched bandits. Motivated by natural restrictions of recommender systems and e-commerce platforms, we assume that a learning agent observes responses batched in groups over a certain time period. Unlike previous work, we consider a more practically relevant batch-centric scenario of batch learning. We provide a policy-agnostic regret analysis and demonstrate upper and lower bounds for the regret of a candidate policy. Our main theoretical results show that the impact of batch learning can be measured in terms of online behavior. Finally, we demonstrate the consistency of theoretical results by conducting empirical experiments and reflect on the optimal batch size choice.
arXiv commentTo appear at the workshop on the Ecological Theory of Reinforcement Learning, NeurIPS 2021
References in corpus (7)
- A Contextual-Bandit Approach to Personalized News Article Recommendation
- Unbiased Offline Evaluation of Contextual-bandit-based News Article Recommendation Algorithms
- Batched bandit problems
- An Efficient Bandit Algorithm for Realtime Multivariate Optimization
- Bandits with Delayed, Aggregated Anonymous Feedback
- Sequential Batch Learning in Finite-Action Linear Contextual Bandits
- Online Learning under Delayed Feedback