Unbiased Implicit Feedback via Bi-level Optimization

TL;DR

This paper introduces a low-variance unbiased estimator for implicit feedback in recommender systems, using bi-level optimization and an unbiased validation set to improve relevance estimation and model performance.

Contribution

It proposes a novel low-variance unbiased estimator and a bi-level optimization framework leveraging an unbiased validation set for implicit feedback modeling.

Findings

Effective in reducing gradient variance for long-tail items

Improves relevance estimation accuracy in recommender systems

Validated on multiple real-world and semi-synthetic datasets

Abstract

Implicit feedback is widely leveraged in recommender systems since it is easy to collect and provides weak supervision signals. Recent works reveal a huge gap between the implicit feedback and user-item relevance due to the fact that implicit feedback is also closely related to the item exposure. To bridge this gap, existing approaches explicitly model the exposure and propose unbiased estimators to improve the relevance. Unfortunately, these unbiased estimators suffer from the high gradient variance, especially for long-tail items, leading to inaccurate gradient updates and degraded model performance. To tackle this challenge, we propose a low-variance unbiased estimator from a probabilistic perspective, which effectively bounds the variance of the gradient. Unlike previous works which either estimate the exposure via heuristic-based strategies or use a large biased training set, we propose to estimate the exposure via an unbiased small-scale validation set. Specifically, we first parameterize the user-item exposure by incorporating both user and item information, and then construct an unbiased validation set from the biased training set. By leveraging the unbiased validation set, we adopt bi-level optimization to automatically update exposure-related parameters along with recommendation model parameters during the learning. Experiments on two real-world datasets and two semi-synthetic datasets verify the effectiveness of our method.