Addressing Bias and Fairness in Search Systems

Gao, Ruoyuan; Shah, Chirag

doi:10.1145/3404835.3462807

Cited by 15 publications

(10 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First of all, pre-processing methods usually aim to minimize the bias in data as bias may arise from the data source. This includes fairness-aware sampling methodologies in the data collection process to cover items of all groups, or balancing methodologies to increase coverage of minority groups, or repairing methodologies to ensure label correctness, remove disparate impact [14]. However, most of the time, we do not have access to the data collection process, but are given the dataset.…”

Section: Related Work 21 Fairness In Recommendationmentioning

confidence: 99%

“…The relevant methods related to fairness in ranking and recommendation can be roughly divided into three categories: preprocessing, in-processing and post-processing algorithms [14,28,29]. First of all, pre-processing methods usually aim to minimize the bias in data as bias may arise from the data source.…”

Section: Related Work 21 Fairness In Recommendationmentioning

confidence: 99%

“…In the area of fairness-aware recommendation, the methods can be roughly divided into three categories: pre-processing, inprocessing and post-processing algorithms [14,29]. Pre-processing methods usually aim to remove bias in data, e.g., sampling from data to cover items of all groups or balancing data to increase coverage of minority groups.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Toward Pareto Efficient Fairness-Utility Trade-off in Recommendation through Reinforcement Learning

Zhao

et al. 2022

Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

View full text Add to dashboard Cite

The issue of fairness in recommendation is becoming increasingly essential as Recommender Systems (RS) touch and influence more and more people in their daily lives. In fairness-aware recommendation, most of the existing algorithmic approaches mainly aim at solving a constrained optimization problem by imposing a constraint on the level of fairness while optimizing the main recommendation objective, e.g., click through rate (CTR). While this alleviates the impact of unfair recommendations, the expected return of an approach may significantly compromise the recommendation accuracy due to the inherent trade-off between fairness and utility. This motivates us to deal with these conflicting objectives and explore the optimal trade-off between them in recommendation. One conspicuous approach is to seek a Pareto efficient/optimal solution to guarantee optimal compromises between utility and fairness. Moreover, considering the needs of real-world e-commerce platforms, it would be more desirable if we can generalize the whole Pareto Frontier, so that the decision-makers can specify any preference of one objective over another based on their current business needs. Therefore, in this work, we propose a fairness-aware recommendation framework using multi-objective reinforcement learning (MORL), called MoFIR (pronounced "more fair"), which is able to learn a single parametric representation for optimal recommendation policies over the space of all possible preferences. Specially, we modify traditional Deep Deterministic Policy Gradient (DDPG) by introducing conditioned network (CN) into it, which conditions the networks directly on these preferences and outputs Q-value-vectors. Experiments on several real-world recommendation datasets verify the superiority of our framework on both fairness metrics and recommendation measures when compared with all other baselines. We also extract the approximate Pareto Frontier on real-world datasets generated by MoFIR and compare to state-of-the-art fairness methods.

show abstract

Section: Related Work 21 Fairness In Recommendationmentioning

confidence: 99%

Section: Related Work 21 Fairness In Recommendationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Toward Pareto Efficient Fairness-Utility Trade-off in Recommendation through Reinforcement Learning

Zhao

et al. 2022

Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

View full text Add to dashboard Cite

show abstract

“…First of all, pre-processing methods usually aim to minimize the bias in the data sources. It includes fairness-aware sampling methodologies in the data collection process to cover items of all groups, balancing methodologies to increase coverage of minority groups, and repairing methodologies to ensure label correctness [23]. Secondly, in-processing methods aim at encoding fairness as part of the objective function, typically as a regularizer [1,6,24,41].…”

Section: Fairness In Recommendationmentioning

confidence: 99%

Explainable Fairness in Recommendation

Ge,

Tan,

Zhu

et al. 2022

Preprint

View full text Add to dashboard Cite

Existing research on fairness-aware recommendation has mainly focused on the quantification of fairness and the development of fair recommendation models, neither of which studies a more substantial problem-identifying the underlying reason of model disparity in recommendation. This information is critical for recommender system designers to understand the intrinsic recommendation mechanism and provides insights on how to improve model fairness to decision makers. Fortunately, with the rapid development of Explainable AI, we can use model explainability to gain insights into model (un)fairness. In this paper, we study the problem of explainable fairness, which helps to gain insights about why a system is fair or unfair, and guides the design of fair recommender systems with a more informed and unified methodology. Particularly, we focus on a common setting with feature-aware recommendation and exposure unfairness, but the proposed explainable fairness framework is general and can be applied to other recommendation settings and fairness definitions. We propose a Counterfactual Explainable Fairness framework, called CEF, which generates explanations about model fairness that can improve the fairness without significantly hurting the performance. The CEF framework formulates an optimization problem to learn the "minimal" change of the input features that changes the recommendation results to a certain level of fairness. Based on the counterfactual recommendation result of each feature, we calculate an explainability score in terms of the fairness-utility trade-off to rank all the feature-based explanations, and select the top ones as fairness explanations. Experimental results on several real-world datasets validate that our method is able to effectively provide explanations to the model disparities and these explanations can achieve better fairness-utility trade-off when using them for recommendation than all the baselines. CCS CONCEPTS• Computing methodologies → Artificial intelligence.

show abstract

“…Even models developed with the best intentions may introduce discriminatory biases. Researchers in various fields have found the unfairness issues such as model compression [Bagdasaryan et al, 2019], differential privacy [Hooker et al, 2019[Hooker et al, , 2020, recommendation system [Gómez et al, 2021], information retrieval [Gao and Shah, 2021], machine translation [Khan et al, 2021], and learning with noisy labels [Liu, 2021]. To our best knowledge, the unfairness issue in SSL methods has not been sufficiently explored.…”

Section: Related Workmentioning

confidence: 99%

The Rich Get Richer: Disparate Impact of Semi-Supervised Learning

Zhu¹,

Luo²,

Liu³

2021

Preprint

View full text Add to dashboard Cite

Semi-supervised learning (SSL) has demonstrated its potential to improve the model accuracy for a variety of learning tasks when the high-quality supervised data is severely limited. Although it is often established that the average accuracy for the entire population of data is improved, it is unclear how SSL fares with different sub-populations. Understanding the above question has substantial fairness implications when these different sub-populations are defined by the demographic groups we aim to treat fairly. In this paper, we reveal the disparate impacts of deploying SSL: the sub-population who has a higher baseline accuracy without using SSL (the "rich" sub-population) tends to benefit more from SSL; while the sub-population who suffers from a low baseline accuracy (the "poor" sub-population) might even observe a performance drop after adding the SSL module. We theoretically and empirically establish the above observation for a broad family of SSL algorithms, which either explicitly or implicitly use an auxiliary "pseudo-label". Our experiments on a set of image and text classification tasks confirm our claims. We discuss how this disparate impact can be mitigated and hope that our paper will alarm the potential pitfall of using SSL and encourage a multifaceted evaluation of future SSL algorithms. Code is available at github.com/UCSC-REAL/Disparate-SSL.

show abstract

Addressing Bias and Fairness in Search Systems

Cited by 15 publications

References 25 publications

Toward Pareto Efficient Fairness-Utility Trade-off in Recommendation through Reinforcement Learning

Toward Pareto Efficient Fairness-Utility Trade-off in Recommendation through Reinforcement Learning

Explainable Fairness in Recommendation

The Rich Get Richer: Disparate Impact of Semi-Supervised Learning

Contact Info

Product

Resources

About