A Survey on Causal Inference for Recommendation
Huishi Luo, Fuzhen Zhuang, Ruobing Xie, Hengshu Zhu, Deqing Wang, Zhulin An, Yongjun Xu
TL;DR
This survey addresses how causal inference can be integrated into recommender systems through a theory-driven lens that unifies potential outcome (PO) and structural causal model (SCM) frameworks, as well as general counterfactual reasoning. It proposes a three-way taxonomy—PO-based, SCM-based, and general counterfactuals-based methods—and systematically maps representative RS approaches to each causal theory, offering a coherent view of 120+ papers and open-source resources. Key contributions include a coherent explanation of when to use propensity-score techniques versus causal-effect estimation, as well as detailed coverage of collider, mediator, and confounder structures, with practical interventions like back-door, front-door, and instrumental-variable strategies. The survey emphasizes practical impact on offline/online policy evaluation, debiasing, uplift modeling, data augmentation, fairness, and explainability, and outlines future directions in causal discovery, transfer learning, dynamic modeling, and causality-aware foundation models for RS.
Abstract
Causal inference has recently garnered significant interest among recommender system (RS) researchers due to its ability to dissect cause-and-effect relationships and its broad applicability across multiple fields. It offers a framework to model the causality in recommender systems like confounding effects and deal with counterfactual problems such as offline policy evaluation and data augmentation. Although there are already some valuable surveys on causal recommendations, they typically classify approaches based on the practical issues faced in RS, a classification that may disperse and fragment the unified causal theories. Considering RS researchers' unfamiliarity with causality, it is necessary yet challenging to comprehensively review relevant studies from a coherent causal theoretical perspective, thereby facilitating a deeper integration of causal inference in RS. This survey provides a systematic review of up-to-date papers in this area from a causal theory standpoint and traces the evolutionary development of RS methods within the same causal strategy. Firstly, we introduce the fundamental concepts of causal inference as the basis of the following review. Subsequently, we propose a novel theory-driven taxonomy, categorizing existing methods based on the causal theory employed - namely, those based on the potential outcome framework, the structural causal model, and general counterfactuals. The review then delves into the technical details of how existing methods apply causal inference to address particular recommender issues. Finally, we highlight some promising directions for future research in this field. Representative papers and open-source resources will be progressively available at https://github.com/Chrissie-Law/Causal-Inference-for-Recommendation.