Combinatorial Causal Bandits without Graph Skeleton

arXiv2301.13392MaRDI QIDQ6424921

Author name not available (Why is that?)

Publication date: 30 January 2023

Abstract: In combinatorial causal bandits (CCB), the learning agent chooses a subset of variables in each round to intervene and collects feedback from the observed variables to minimize expected regret or sample complexity. Previous works study this problem in both general causal models and binary generalized linear models (BGLMs). However, all of them require prior knowledge of causal graph structure. This paper studies the CCB problem without the graph structure on binary general causal models and BGLMs. We first provide an exponential lower bound of cumulative regrets for the CCB problem on general causal models. To overcome the exponentially large space of parameters, we then consider the CCB problem on BGLMs. We design a regret minimization algorithm for BGLMs even without the graph skeleton and show that it still achieves $O(sqrtTlnT)$ expected regret. This asymptotic regret is the same as the state-of-art algorithms relying on the graph structure. Moreover, we sacrifice the regret to $O(T^{frac23}lnT)$ to remove the weight gap covered by the asymptotic notation. At last, we give some discussions and algorithms for pure exploration of the CCB problem without the graph structure.

Has companion code repository: https://github.com/fengtony686/CCB-without-Graph

This page was built for publication: Combinatorial Causal Bandits without Graph Skeleton

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6424921)