# Using causal abstractions to accelerate decision-making in complex bandit problems

arXiv: https://arxiv.org/abs/2509.04296 · HTML: https://mlsanity.com/paper/2509.04296/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 27 references that are in the map.
- First posted on arXiv 4 Sep 2025.

## Builds on (most cited first, 12 of 12)

- [Causal Consistency of Structural Equation Models](https://mlsanity.com/paper/1707.00819/) (2017)
- [Abstracting Causal Models](https://mlsanity.com/paper/1812.03789/) (2018)
- [Causal Bandits: Learning Good Interventions via Causal Inference](https://mlsanity.com/paper/1606.03203/) (2016)
- [Adaptively Exploiting d-Separators with Causal Bandits](https://mlsanity.com/paper/2202.05100/) (2022)
- [Multi-Armed Bandits with Dependent Arms](https://mlsanity.com/paper/2010.09478/) (2020)
- [Online Bandits with (Biased) Offline Data: Adaptive Learning under Distribution Mismatch](https://mlsanity.com/paper/2405.02594/) (2024)
- [Quantifying Consistency and Information Loss for Causal Abstraction Learning](https://mlsanity.com/paper/2305.04357/) (2023)
- [Regional Multi-Armed Bandits](https://mlsanity.com/paper/1802.07917/) (2018)
- [Causally Abstracted Multi-armed Bandits](https://mlsanity.com/paper/2404.17493/) (2024)
- [Aligning Graphical and Functional Causal Abstractions](https://mlsanity.com/paper/2412.17080/) (2024)
- [Best Arm Identification with Possibly Biased Offline Data](https://mlsanity.com/paper/2505.23165/) (2025)
- [Graph Feedback Bandits with Similar Arms](https://mlsanity.com/paper/2405.11171/) (2024)

## Built on by (most cited first, 1 of 1)

- [Coarsening Causal DAG Models](https://mlsanity.com/paper/2601.10531/) (2026)

## Abstract (arXiv)

Although real-world decision-making problems can often be encoded as causal multi-armed bandits (CMABs) at different levels of abstraction, a general methodology exploiting the information and computational advantages of each abstraction level is missing. In this paper, we propose AT-UCB, an algorithm which efficiently exploits shared information between CMAB problem instances defined at different levels of abstraction. More specifically, AT-UCB leverages causal abstraction (CA) theory to explore within a cheap-to-simulate and coarse-grained CMAB instance, before employing the traditional upper confidence bound (UCB) algorithm on a restricted set of potentially optimal actions in the CMAB of interest, leading to significant reductions in cumulative regret when compared to the classical UCB algorithm. We illustrate the advantages of AT-UCB theoretically, through a novel upper bound on the cumulative regret, and empirically, by applying AT-UCB to epidemiological simulators with varying resolution and computational cost.