A Novel Generative Model with Causality Constraint for Mitigating Biases in Recommender Systems

TL;DR

This paper introduces LCDR, a generative model with causality constraints that effectively mitigates bias in recommender systems by aligning latent representations, even with weak proxy variables, leading to improved recommendation accuracy.

Contribution

The paper proposes a novel causal generative framework using an identifiable VAE to better recover latent confounders in recommender systems, addressing limitations of existing methods.

Findings

LCDR outperforms existing bias mitigation methods in experiments.

The model effectively leverages noisy proxy variables to recover confounders.

Improved recommendation accuracy demonstrated on real-world datasets.

Abstract

Accurately predicting counterfactual user feedback is essential for building effective recommender systems. However, latent confounding bias can obscure the true causal relationship between user feedback and item exposure, ultimately degrading recommendation performance. Existing causal debiasing approaches often rely on strong assumptions-such as the availability of instrumental variables (IVs) or strong correlations between latent confounders and proxy variables-that are rarely satisfied in real-world scenarios. To address these limitations, we propose a novel generative framework called Latent Causality Constraints for Debiasing representation learning in Recommender Systems (LCDR). Specifically, LCDR leverages an identifiable Variational Autoencoder (iVAE) as a causal constraint to align the latent representations learned by a standard Variational Autoencoder (VAE) through a unified loss function. This alignment allows the model to leverage even weak or noisy proxy variables to recover latent confounders effectively. The resulting representations are then used to improve recommendation performance. Extensive experiments on three real-world datasets demonstrate that LCDR consistently outperforms existing methods in both mitigating bias and improving recommendation accuracy.