# Reconsidering Learning Objectives in Unbiased Recommendation with Unobserved Confounders

arXiv: https://arxiv.org/abs/2206.03851 · HTML: https://mlsanity.com/paper/2206.03851/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 62 references that are in the map.
- First posted on arXiv 7 Jun 2022.

## Builds on (most cited first, 27 of 27)

- [Unsupervised Domain Adaptation by Backpropagation](https://mlsanity.com/paper/1409.7495/) (2014)
- [MixMatch: A Holistic Approach to Semi-Supervised Learning](https://mlsanity.com/paper/1905.02249/) (2019)
- [Neural Collaborative Filtering](https://mlsanity.com/paper/1708.05031/) (2017)
- [f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization](https://mlsanity.com/paper/1606.00709/) (2016)
- [Learning from Simulated and Unsupervised Images through Adversarial Training](https://mlsanity.com/paper/1612.07828/) (2016)
- [Recommendations as Treatments: Debiasing Learning and Evaluation](https://mlsanity.com/paper/1602.05352/) (2016)
- [Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data](https://mlsanity.com/paper/2010.03622/) (2020)
- [Support and Invertibility in Domain-Invariant Representations](https://mlsanity.com/paper/1903.03448/) (2019)
- [Doubly robust off-policy evaluation with shrinkage](https://mlsanity.com/paper/1907.09623/) (2019)
- [Causal Embeddings for Recommendation](https://mlsanity.com/paper/1706.07639/) (2017)
- [Unbiased Recommender Learning from Missing-Not-At-Random Implicit Feedback](https://mlsanity.com/paper/1909.03601/) (2019)
- [Maximum Density Divergence for Domain Adaptation](https://mlsanity.com/paper/2004.12615/) (2020)
- [IRGAN: A Minimax Game for Unifying Generative and Discriminative Information Retrieval Models](https://mlsanity.com/paper/1705.10513/) (2017)
- [Self-training Avoids Using Spurious Features Under Domain Shift](https://mlsanity.com/paper/2006.10032/) (2020)
- [AutoDebias: Learning to Debias for Recommendation](https://mlsanity.com/paper/2105.04170/) (2021)
- [Off-policy Bandits with Deficient Support](https://mlsanity.com/paper/2006.09438/) (2020)
- [Adversarial Personalized Ranking for Recommendation](https://mlsanity.com/paper/1808.03908/) (2018)
- [FairRec: Fairness-aware News Recommendation with Decomposed Adversarial Learning](https://mlsanity.com/paper/2006.16742/) (2020)
- [DARec: Deep Domain Adaptation for Cross-Domain Recommendation via Transferring Rating Patterns](https://mlsanity.com/paper/1905.10760/) (2019)
- [ESAM: Discriminative Domain Adaptation with Non-Displayed Items to Improve Long-Tail Performance](https://mlsanity.com/paper/2005.10545/) (2020)
- [Information Theoretic Counterfactual Learning from Missing-Not-At-Random Feedback](https://mlsanity.com/paper/2009.02623/) (2020)
- [Asymmetric Tri-training for Debiasing Missing-Not-At-Random Explicit Feedback](https://mlsanity.com/paper/1910.01444/) (2019)
- [Enhanced Doubly Robust Learning for Debiasing Post-click Conversion Rate Estimation](https://mlsanity.com/paper/2105.13623/) (2021)
- [Correcting Exposure Bias for Link Recommendation](https://mlsanity.com/paper/2106.07041/) (2021)
- [Adversarial Counterfactual Learning and Evaluation for Recommender System](https://mlsanity.com/paper/2012.02295/) (2020)
- [Rethinking Neural vs. Matrix-Factorization Collaborative Filtering: the Theoretical Perspectives](https://mlsanity.com/paper/2110.12141/) (2021)
- [Cross Pairwise Ranking for Unbiased Item Recommendation](https://mlsanity.com/paper/2204.12176/) (2022)

## Abstract (arXiv)

This work studies the problem of learning unbiased algorithms from biased feedback for recommendation. We address this problem from a novel distribution shift perspective. Recent works in unbiased recommendation have advanced the state-of-the-art with various techniques such as re-weighting, multi-task learning, and meta-learning. Despite their empirical successes, most of them lack theoretical guarantees, forming non-negligible gaps between theories and recent algorithms. In this paper, we propose a theoretical understanding of why existing unbiased learning objectives work for unbiased recommendation. We establish a close connection between unbiased recommendation and distribution shift, which shows that existing unbiased learning objectives implicitly align biased training and unbiased test distributions. Built upon this connection, we develop two generalization bounds for existing unbiased learning methods and analyze their learning behavior. Besides, as a result of the distribution shift, we further propose a principled framework, Adversarial Self-Training (AST), for unbiased recommendation. Extensive experiments on real-world and semi-synthetic datasets demonstrate the effectiveness of AST.