Online Policy Learning and Inference by Matrix Completion

arXiv:2404.17398v1 Announce Type: cross
Abstract: Making online decisions can be challenging when features are sparse and orthogonal to historical ones, especially when the optimal policy is learned through collaborative filtering. We formulate the problem as a matrix completion bandit (MCB), where the expected reward under each arm is characterized by an unknown low-rank matrix. The $\epsilon$-greedy bandit and the online gradient descent algorithm are explored. Policy learning and regret performance are studied under a specific schedule for exploration probabilities and …

abstract arm arxiv collaborative collaborative filtering cs.lg decisions features filtering inference making matrix ones policy stat.ml through type