New Fairness Metrics for Recommendation that Embrace Differences

By Sirui Yao, Bert Huang. 2017

39 citations recorded by the source corpus.

We study fairness in collaborative-filtering recommender systems, which are sensitive to discrimination that exists in historical data. Biased data can lead collaborative filtering methods to make unfair predictions against minority groups of users. We identify the insufficiency of existing fairness metrics and propose four new metrics that address different forms of unfairness. These fairness metrics can be optimized by adding fairness terms to the learning objective. Experiments on synthetic and real data show that our new metrics can better measure fairness than the baseline, and that the fairness objectives effectively help reduce unfairness.

Subfields: Artificial Intelligence, Information Systems, Safety Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiangnan He
2. Xiang Wang
3. Fuli Feng
4. Meng Wang
5. Emma Brunskill
6. Tat-seng Chua
7. Philip S. Thomas
8. Bruno Castro da Silva

In Information Systems

1. Xiangnan He
2. Xiang Wang
3. Fuli Feng
4. Meng Wang
5. Dietmar Jannach
6. Jiawei Chen
7. Tat-seng Chua
8. Gediminas Adomavicius

In Safety Research

1. Dana Pessach
2. Erez Shmueli
3. Luiz Augusto Sangoi Pizzato
4. Jiawei Chen
5. Dietmar Jannach
6. Andrew Barto
7. Gediminas Adomavicius
8. Robin D. Burke

Citations in the corpus

• Compositional Fairness Constraints for Graph Embeddings (2019, background): “…on general classification models (Chouldechova, 2017; Gajane & Pechenizkiy, 2017; Kamishima et al., 2012) and collaborative filtering algorithms (Yao & Huang, 2017) has received considerable attention in recent years, these techniques have yet to be considered within the context of graph…”
• Algorithmic Fairness (2020, background): “[140]and [78] suggest an in-process mechanism by including additional regularization factors in the objective functions.”
• Bias and Debias in Recommender System: A Survey and Future Directions (2020, methodology): “[127], [128] proposed four fairness metrics in collaborative filtering, and used similar regularization-based optimization method to mitigate different forms of bias.”
• Are My Deep Learning Systems Fair? An Empirical Study of Fixed-Seed Training (2021, background): “Bias metrics: Many metrics [6, 15, 19, 24, 26, 31, 35, 44, 46, 67, 82, 102, 103] have been proposed to measure biases of ML systems.”
• Multistakeholder recommendation: Survey and research directions (2020, methodology): “The error independence criterion was approached through regularization in Yao and Huang (2017a).”
• A Review on Fairness in Machine Learning (2022, background): “Yao and Huang [212] and Kamishima et al. [113] suggest an in-process mechanism by including additional regularization factors in the objective functions.”
• Preventing undesirable behavior of intelligent machines (2019, background): “Finally, other work has introduced definitions of fairness that are tailored to specific applications, such as for use in evaluating recommendation systems [127].”