Contextual Bandit with Missing Rewards
arXiv:2007.06368
Abstract
We consider a novel variant of the contextual bandit problem (i.e., the multi-armed bandit with side-information, or context, available to a decision-maker) where the reward associated with each context-based decision may not always be observed("missing rewards"). This new problem is motivated by certain online settings including clinical trial and ad recommendation applications. In order to address the missing rewards setting, we propose to combine the standard contextual bandit approach with an unsupervised learning mechanism such as clustering. Unlike standard contextual bandit methods, by leveraging clustering to estimate missing reward, we are able to learn from each incoming event, even those with missing rewards. Promising empirical results are obtained on several real-life datasets.
References in corpus (10)
- A Contextual-Bandit Approach to Personalized News Article Recommendation
- Thompson Sampling for Contextual Bandits with Linear Payoffs
- A Survey on Practical Applications of Multi-Armed and Contextual Bandits
- Beyond Backprop: Online Alternating Minimization with Auxiliary Variables
- Online Semi-Supervised Learning with Deep Hybrid Boltzmann Machines and Denoising Autoencoders
- Interpretable Multi-Objective Reinforcement Learning through Policy Orchestration
- Unified Models of Human Behavioral Agents in Bandits, Contextual Bandits and RL
- A Bandit Approach to Posterior Dialog Orchestration Under a Budget
- Online learning with Corrupted context: Corrupted Contextual Bandits
- Hyper-parameter Tuning for the Contextual Bandit