Improved Bound for Robust Causal Bandits with Linear Models
This work addresses the problem of robust causal bandits for researchers and practitioners dealing with temporal model fluctuations in causal systems, providing an incremental improvement over existing methods.
The authors tackled the problem of robust causal bandits with linear models, achieving an upper bound of $ ilde{mathcal{O}}(d^{L- rac{1}{2}}(sqrt{T} + C))$ and a lower bound of $Ω(d^{ rac{L}{2}-2}max{sqrt{T}; ,; d^2C})$ on cumulative regret. The proposed algorithm achieves nearly optimal $ ilde{mathcal{O}}(sqrt{T})$ regret when $C$ is $o(sqrt{T})$.
This paper investigates the robustness of causal bandits (CBs) in the face of temporal model fluctuations. This setting deviates from the existing literature's widely-adopted assumption of constant causal models. The focus is on causal systems with linear structural equation models (SEMs). The SEMs and the time-varying pre- and post-interventional statistical models are all unknown and subject to variations over time. The goal is to design a sequence of interventions that incur the smallest cumulative regret compared to an oracle aware of the entire causal model and its fluctuations. A robust CB algorithm is proposed, and its cumulative regret is analyzed by establishing both upper and lower bounds on the regret. It is shown that in a graph with maximum in-degree $d$, length of the largest causal path $L$, and an aggregate model deviation $C$, the regret is upper bounded by $\tilde{\mathcal{O}}(d^{L-\frac{1}{2}}(\sqrt{T} + C))$ and lower bounded by $Ω(d^{\frac{L}{2}-2}\max\{\sqrt{T}\; ,\; d^2C\})$. The proposed algorithm achieves nearly optimal $\tilde{\mathcal{O}}(\sqrt{T})$ regret when $C$ is $o(\sqrt{T})$, maintaining sub-linear regret for a broad range of $C$.