ABSTRACT
Many internet applications are powered by machine learned models, which are usually trained on labeled datasets obtained through user feedback signals or human judgments. Since societal biases may be present in the generation of such datasets, it is possible for the trained models to be biased, thereby resulting in potential discrimination and harms for disadvantaged groups. Motivated by the need to understand and address algorithmic bias in web-scale ML systems and the limitations of existing fairness toolkits, we present the LinkedIn Fairness Toolkit (LiFT), a framework for scalable computation of fairness metrics as part of large ML systems. We highlight the key requirements in deployed settings, and present the design of our fairness measurement system. We discuss the challenges encountered in incorporating fairness tools in practice and the lessons learned during deployment at LinkedIn. Finally, we provide open problems based on practical experience.
Supplemental Material
- M. Abadi et al. TensorFlow: A system for large-scale machine learning. In OSDI, 2016.Google Scholar
- J. A. Adebayo. FairML: ToolBox for diagnosing bias in predictive modeling. PhD thesis, Massachusetts Institute of Technology, 2016.Google Scholar
- A. Agarwal, A. Beygelzimer, M. Dudík, J. Langford, and H.Wallach. A reductions approach to fair classification. In ICML, 2018.Google Scholar
- M. Armbrust et al. Spark SQL: Relational data processing in Spark. In SIGMOD, 2015.Google Scholar
- N. Bantilan. Themis-ML: A fairness-aware machine learning interface for end-toend discrimination discovery and mitigation. Journal of Technology in Human Services, 36(1), 2018.Google ScholarCross Ref
- S. Barocas and M. Hardt. Fairness in machine learning. In NeurIPS Tutorial, 2017.Google Scholar
- R. K. Bellamy et al. AI Fairness 360: An extensible toolkit for detecting, understanding, and mitigating unwanted algorithmic bias. arXiv preprint arXiv:1810.01943, 2018.Google Scholar
- A. Beutel, J. Chen, T. Doshi, H. Qian, A.Woodruff, C. Luu, P. Kreitmann, J. Bischof, and E. H. Chi. Putting fairness principles into practice: Challenges, metrics, and improvements. In AIES, 2019.Google ScholarDigital Library
- A. Chouldechova. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Big data, 5(2), 2017.Google Scholar
- K. Crawford. The trouble with bias. In NeurIPS Invited Talk, 2017.Google Scholar
- C. DiCiccio, S. Vasudevan, K. Basu, K. Kenthapadi, and D. Agarwal. Evaluating fairness using permutation tests. In KDD, 2020.Google ScholarDigital Library
- S. A. Friedler, C. Scheidegger, S. Venkatasubramanian, S. Choudhary, E. P. Hamilton, and D. Roth. A comparative study of fairness-enhancing interventions in machine learning. In FAT*, 2019.Google ScholarDigital Library
- S. Galhotra, Y. Brun, and A. Meliou. Fairness testing: Testing software for discrimination. In ESEC/FSE, 2017.Google ScholarDigital Library
- S. C. Geyik, S. Ambler, and K. Kenthapadi. Fairness-aware ranking in search & recommendation systems with application to LinkedIn talent search. In KDD, 2019.Google ScholarDigital Library
- P. Good. Permutation Tests: A Practical Guide to Resampling Methods for Testing Hypotheses. Springer series in statistics. Springer, 2000.Google ScholarCross Ref
- B. Green and Y. Chen. Disparate interactions: An algorithm-in-the-loop analysis of fairness in risk assessments. In FAT*, 2019.Google ScholarDigital Library
- N. Grgic-Hlaca, E. M. Redmiles, K. P. Gummadi, and A. Weller. Human perceptions of fairness in algorithmic decision making: A case study of criminal risk prediction. In WWW, 2018.Google Scholar
- M. Hardt, E. Price, and N. Srebro. Equality of opportunity in supervised learning. In NIPS, 2016.Google ScholarDigital Library
- K. Holstein, J. Wortman Vaughan, H. Daumé III, M. Dudik, and H. Wallach. Improving fairness in machine learning systems: What do industry practitioners need? In CHI, 2019.Google ScholarDigital Library
- F. Kamiran and T. Calders. Data preprocessing techniques for classification without discrimination. Knowledge and Information Systems, 33(1):1--33, 2012.Google ScholarDigital Library
- T. Kamishima, S. Akaho, H. Asoh, and J. Sakuma. Fairness-aware classifier with prejudice remover regularizer. In ECML PKDD, 2012.Google ScholarCross Ref
- R. Kohavi. Scaling up the accuracy of Naive-Bayes classifiers: A decision-tree hybrid. In KDD, 1996.Google ScholarDigital Library
- M. Mitchell, S.Wu, A. Zaldivar, P. Barnes, L. Vasserman, B. Hutchinson, E. Spitzer, I. D. Raji, and T. Gebru. Model cards for model reporting. In FAT*, 2019.Google ScholarDigital Library
- M. Ojala and G. C. Garriga. Permutation tests for studying classifier performance. Journal of Machine Learning Research, 11, 2010.Google Scholar
- F. Pedregosa et al. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2011.Google Scholar
- pymetrics. audit-AI, 2018. https://github.com/pymetrics/audit-ai.Google Scholar
- P. Saleiro, B. Kuester, A. Stevens, A. Anisfeld, L. Hinkson, J. London, and R. Ghani. Aequitas: A bias and fairness audit toolkit. arXiv preprint arXiv:1811.05577, 2018.Google Scholar
- T. Speicher, H. Heidari, N. Grgic-Hlaca, K. P. Gummadi, A. Singla, A. Weller, and M. B. Zafar. A unified approach to quantifying algorithmic unfairness: Measuring individual & group unfairness via inequality indices. In KDD, 2018.Google ScholarDigital Library
- M. Srivastava, H. Heidari, and A. Krause. Mathematical notions vs. human perception of fairness: A descriptive approach to fairness for machine learning. In KDD, 2019.Google ScholarDigital Library
- A. Swami, S. Vasudevan, and J. Huyn. Data sentinel: A declarative production scale data validation platform. In ICDE, 2020.Google ScholarCross Ref
- F. Tramer, V. Atlidakis, R. Geambasu, D. Hsu, J.-P. Hubaux, M. Humbert, A. Juels, and H. Lin. FairTest: Discovering unwarranted associations in data-driven applications. In Euro S&P, 2017.Google ScholarCross Ref
- M. Zaharia et al. Apache spark: a unified engine for big data processing. Communications of the ACM, 59(11), 2016.Google ScholarDigital Library
- M. Zehlike, C. Castillo, F. Bonchi, S. Hajian, and M. Megahed. Fairness Measures: Datasets and software for detecting algorithmic discrimination, 2017. http://fairness-measures.org/.Google Scholar
Index Terms
- LiFT: A Scalable Framework for Measuring Fairness in ML Applications
Recommendations
A Review on Fairness in Machine Learning
An increasing number of decisions regarding the daily lives of human beings are being controlled by artificial intelligence and machine learning (ML) algorithms in spheres ranging from healthcare, transportation, and education to college admissions, ...
Fairness-Aware Machine Learning: Practical Challenges and Lessons Learned
WSDM '19: Proceedings of the Twelfth ACM International Conference on Web Search and Data MiningResearchers and practitioners from different disciplines have highlighted the ethical and legal challenges posed by the use of machine learned models and data-driven systems, and the potential for such systems to discriminate against certain population ...
Bias Mitigation for Machine Learning Classifiers: A Comprehensive Survey
This paper provides a comprehensive survey of bias mitigation methods for achieving fairness in Machine Learning (ML) models. We collect a total of 341 publications concerning bias mitigation for ML classifiers. These methods can be distinguished based on ...
Comments