Mitigating Confounding Bias in Recommendation via Information Bottleneck

Liu, Dugang; Cheng, Pengxiang; Zhu, Hong; Dong, Zhenhua; He, Xiuqiang; Pan, Weike; Ming, Zhong

doi:10.1145/3460231.3474263

Cited by 73 publications

(21 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[6,29] adjust data distribution to estimate causal effect with IPW methods. [20] solves the confounding problem with information bottleneck [20]. These methods do not perform intervention with do-calculus.…”

Section: Causal Methods For Recommendationmentioning

confidence: 99%

Addressing Confounding Feature Issue for Causal Recommendation

He¹,

Zhang²,

Feng³

et al. 2022

Preprint

View full text Add to dashboard Cite

In recommender system, some feature directly affects whether an interaction would happen, making the happened interactions not necessarily indicate user preference. For instance, short videos are objectively easier to be finished even though the user does not like the video. We term such feature as confounding feature, and video length is a confounding feature in video recommendation. If we fit a model on such interaction data, just as done by most data-driven recommender systems, the model will be biased to recommend short videos more, and deviate from user actual requirement.This work formulates and addresses the problem from the causal perspective. Assuming there are some factors affecting both the confounding feature and other item features, e.g., the video creator, we find the confounding feature opens a backdoor path behind user-item matching and introduces spurious correlation. To remove the effect of backdoor path, we propose a framework named Deconfounding Causal Recommendation (DCR), which performs intervened inference with do-calculus. Nevertheless, evaluating do-calculus requires to sum over the prediction on all possible values of confounding feature, significantly increasing the time cost. To address the efficiency challenge, we further propose a mixture-of-experts (MoE) model architecture, modeling each value of confounding feature with a separate expert module. Through this way, we retain the model expressiveness with few additional costs. We demonstrate DCR on the backbone model of neural factorization machine (NFM), showing that DCR leads to more accurate prediction of user preference with small inference time cost.

show abstract

Section: Causal Methods For Recommendationmentioning

confidence: 99%

Addressing Confounding Feature Issue for Causal Recommendation

He¹,

Zhang²,

Feng³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…MACR [37] and CR [34] removes the direct effect of item properties on predicted scores by causal inference. DIB [25] and PDA [43] both remove the confounding popularity bias during training, but PDA [43] further inject the future popularity to the scores during inference. Unlike our proposed method that focuses on constructing an unbiased loss function, these approaches deal with the popularity bias from a different point of view -they analyze the causal effect between the bias and the observed data and then apply causal operations accordingly.…”

Section: Related Workmentioning

confidence: 99%

Cross Pairwise Ranking for Unbiased Item Recommendation

Wan

Wang

et al. 2022

Proceedings of the ACM Web Conference 2022

View full text Add to dashboard Cite

Most recommender systems optimize the model on observed interaction data, which is affected by the previous exposure mechanism and exhibits many biases like popularity bias. The loss functions, such as the mostly used pointwise Binary Cross-Entropy and pairwise Bayesian Personalized Ranking, are not designed to consider the biases in observed data. As a result, the model optimized on the loss would inherit the data biases, or even worse, amplify the biases. For example, a few popular items take up more and more exposure opportunities, severely hurting the recommendation quality on niche items -known as the notorious Mathew effect.In this work, we develop a new learning paradigm named Cross Pairwise Ranking (CPR) that achieves unbiased recommendation without knowing the exposure mechanism. Distinct from inverse propensity scoring (IPS), we change the loss term of a sample -we innovatively sample multiple observed interactions once and form the loss as the combination of their predictions. We prove in theory that this way offsets the influence of user/item propensity on the learning, removing the influence of data biases caused by the exposure mechanism. Advantageous to IPS, our proposed CPR ensures unbiased learning for each training instance without the need of setting the propensity scores. Experimental results demonstrate the superiority of CPR over state-of-the-art debiasing solutions in both model generalization and training efficiency. The codes are available at https://github.com/Qcactus/CPR. CCS CONCEPTS• Information systems → Retrieval models and ranking; Recommender systems.

show abstract

“…At present, some causal reasoning works [213][214][215][216][217] has been applied to the recommendation system. The recommendation system is actually a problem of causal reasoning [213].…”

Section: Future Directionsmentioning

confidence: 99%

Causal Reasoning Meets Visual Representation Learning: A Prospective Study

Liu¹,

Yushen²,

Yan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Spatial-temporal representation learning is ubiquitous in various real-world applications, including visual comprehension, video understanding, multi-modal analysis, human-computer interaction, and urban computing. Due to the emergence of huge amounts of multi-modal heterogeneous spatial/temporal/spatial-temporal data in big data era, the lack of interpretability, robustness, and out-of-distribution generalization are becoming the challenges of the existing visual models. The majority of the existing methods tend to fit the original data/variable distributions and ignore the essential causal relations behind the multi-modal knowledge, which lacks an unified guidance and analysis about why modern spatial-temporal representation learning methods are easily collapse into data bias and have limited generalization and cognitive abilities. Inspired by the strong inference ability of human-level agents, recent years have therefore witnessed great effort in developing causal reasoning paradigms to realize robust representation and model learning with good cognitive ability. In this paper, we conduct a comprehensive review of existing causal reasoning methods for spatialtemporal representation learning, covering fundamental theories, models, and datasets. The limitations of current methods and datasets are also discussed. Moreover, we propose some primary challenges, opportunities, and future research directions for benchmarking causal reasoning algorithms in spatialtemporal representation learning. This paper aims to provide a comprehensive overview of this emerging field, attract attention, encourage discussions, bring to the forefront the urgency of developing novel causal reasoning methods, publicly available benchmarks, and consensus-building standards for reliable spatial-temporal representation learning and related real-world applications more efficiently.

show abstract

Mitigating Confounding Bias in Recommendation via Information Bottleneck

Cited by 73 publications

References 28 publications

Addressing Confounding Feature Issue for Causal Recommendation

Addressing Confounding Feature Issue for Causal Recommendation

Cross Pairwise Ranking for Unbiased Item Recommendation

Causal Reasoning Meets Visual Representation Learning: A Prospective Study

Contact Info

Product

Resources

About