Causal Inference for Recommender Systems
Authors:
Open Access
- 22 September 2020
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 426-431
Abstract
The task of recommender systems is classically framed as a prediction of users’ preferences and users’ ratings. However, its spirit is to answer a counterfactual question: “What would the rating be if we ‘forced’ the user to watch the movie?” This is a question about an intervention, that is a causal inference question. The key challenge of this causal inference is unobserved confounders, variables that affect both which items the users decide to interact with and how they rate them. To this end, we develop an algorithm that leverages classical recommendation models for causal recommendation. Across simulated and real datasets, we demonstrate that the proposed algorithm is more robust to unobserved confounders and improves recommendation.Keywords
Funding Information
- FRQNT
- NIH (1U01MH11572701)
- NSF (CCF-1740833)
- Nvidia
- Simons Foundation
- CIFAR AI chair program
- IVADO
- Samsung
- ONR (N00014-17-1-2131,N00014-15-1-2209)
- DARPA (SD2 FA8750-18-C-0130)
- Amazon
- NSERC
This publication has 14 references indexed in Scilit:
- Causal embeddings for recommendation2018
- Causal Inference: A Missing Data PerspectiveStatistical Science, 2018
- Modeling User Exposure in Recommendation2016
- Counterfactual Estimation and Optimization of Click Metrics in Search Engines2015
- Causal Inference for Statistics, Social, and Biomedical Sciences2015
- A contextual-bandit approach to personalized news article recommendation2010
- Collaborative prediction and ranking with non-random missing data2009
- Causality2009
- Collaborative Filtering for Implicit Feedback Datasets2008
- The central role of the propensity score in observational studies for causal effectsBiometrika, 1983