Skip to main content
SpringerLink
Log in

Towards fair machine learning using combinatorial methods

  • Research Paper
  • Published:
Evolutionary Intelligence Aims and scope Submit manuscript

Abstract

With the rise of artificial intelligence and machine learning in the last decade, there has been an increasing interest in developing a solid theory and implementing algorithmic fairness, which has eventually resulted in a large volume of work over the past few years. Despite the enormous amount of work done on the topic over a concise period, there has been little consensus of a unifying theory of algorithmic fairness. In this paper, we develop a notion of fairness that is based on the notion of discrepancy of set systems, a widely studied topic in the theory of computer science and combinatorics. (Chazelle Bernard in The discrepancy method: randomness and complexity. Cambridge University Press (2001)).

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

References

  1. Chazelle Bernard (2001) The discrepancy method: randomness and complexity. Cambridge University Press

  2. Mehrabi Ninareh et al (2021) A survey on bias and fairness in machine learning. ACM Comput Surv (CSUR) 54.6:1–35

  3. Zafar Muhammad Bilal et al (2019) Fairness constraints: a flexible approach for fair classification. J Mach Learn Res 20.1:2737–2778

  4. Barocas S, Hardt M, Narayanan A (2017) Fairness in machine learning. Nips tutorial 1:2017

  5. Patro Gourab K et al (2020) Fairrec: Two-sided fairness for personalized recommendations in two-sided platforms. Proceedings of the web conference

  6. Morik Marco et al (2020) Controlling fairness and bias in dynamic learning-to-rank. Proceedings of the 43rd international ACM SIGIR conference on research and development in information retrieval

  7. Feldman Michael et al (2015) Certifying and removing disparate impact. In: Proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining, pp 259–268

  8. Louizos Christos et al (2017) The variational fair autoencoder

  9. Corbett-Davies S, Sharad Goel (2018) A critical review of fair machine learning, the measure and mismeasure of fairness

  10. Menon Aditya Krishna, Robert C Williamson (2017) The cost of fairness in classification. arXiv preprint arXiv: 1705.09055

  11. Chouldechova A (2017) Fair prediction with disparate impact: a study of bias in recidivism prediction instruments. Big data 5(2):153–163

    Article  Google Scholar 

  12. Kleinberg Jon et al (2018) Algorithmic fairness. Aea papers and proceedings. Vol. 108

  13. Higgins Irina et al (2018) Towards a definition of disentangled representations. arXiv preprint arXiv: 1812.02230

  14. Sattigeri Prasanna et al (2019) Fairness GAN: generating datasets with fairness properties using a generative adversarial network. IBM J Res Dev 63.4/5:3-1

  15. Noriega-Campero Alejandro et al (2019) Active fairness in algorithmic decision making. Proceedings of the 2019 AAAI/ACM conference on AI, ethics, and society

  16. Locatello Francesco et al (2019) On the fairness of disentangled representations. Adv Neural Inf Process Syst 32:14611–14624

  17. Valdivia A, Sánchez-Monedero J, Casillas J (2021) How fair can we go in machine learning? Assessing the boundaries of accuracy and fairness. Int J Intell Syst 36(4):1619–1643

    Article  Google Scholar 

  18. Villar David, Jorge Casillas (2021) Facing many objectives for fairness in machine learning. International conference on the quality of information and communications technology. Springer, Cham

  19. Zhang Yue, Ramesh Arti (2020) Learning fairness-aware relational structures. arXiv preprint arXiv: 2002.09471

  20. Corput JG (1935) van der: Verteilungsfunktionen. Proc. Ned. Akad. v. Wet. Vol 38

  21. Tao Terence (2016) The Erdös discrepancy problem. Discrete Analysis: 609

  22. Spencer J (1985) Six standard deviations suffice. Trans Am Math Soc 289(2):679–706

    Article  MathSciNet  MATH  Google Scholar 

  23. Bansal Nikhil (2010) Constructive algorithms for discrepancy minimization. 2010 IEEE 51st annual symposium on foundations of computer science. IEEE

  24. Lovett S, Meka R (2015) Constructive discrepancy minimization by walking on the edges. SIAM J Comput 44(5):1573–1582

    Article  MathSciNet  MATH  Google Scholar 

  25. Larsen Kasper Green (2019) Constructive discrepancy minimization with hereditary L2 guarantees. 36th International symposium on theoretical aspects of computer science

  26. Charikar Moses, Newman Alantha, Nikolov Aleksandar (2011) Tight hardness results for minimizing discrepancy. Proceedings of the twenty-second annual ACM-SIAM symposium on Discrete Algorithms. Society for Industrial and Applied Mathematics

  27. Kearns Michael et al (2018) Preventing fairness gerrymandering: auditing and learning for subgroup fairness. International conference on machine learning. PMLR

  28. Doerr B, Srivastav A (2003) Multicolour discrepancies. Comb Probab Comput 12(4):365–399

    Article  MathSciNet  MATH  Google Scholar 

  29. Beck J, Fiala T (1981) Integer-making’’ theorems. Discret Appl Math 3(1):1–8

    Article  MathSciNet  MATH  Google Scholar 

  30. Kelkar Shreeharsh (2013) The elite’s last stand: negotiating toughness and fairness in the IIT-JEE, 1990-2005

  31. Propublica Risk Assessment. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing

  32. Adult Dataset. https://archive.ics.uci.edu/ml/datasets/adult

  33. Tabibnia G, Lieberman MD (2007) Fairness and cooperation are rewarding: evidence from social cognitive neuroscience. Ann N Y Acad Sci 1118(1):90–101

    Article  Google Scholar 

  34. Crockett Molly J et al (2008) Serotonin modulates behavioral reactions to unfairness. Science 320.5884:1739–1739

  35. Tabibnia G, Satpute AB, Lieberman MD (2008) The sunny side of fairness: preference for fairness activates reward circuitry (and disregarding unfairness activates self-control circuitry). Psychol Sci 19(4):339–347

    Article  Google Scholar 

Download references

Acknowledgements

This paper is an outcome of the R&D work undertaken project under the Visvesvaraya PhD Scheme of the Ministry of Electronics & Information Technology, Government of India, being implemented by Digital India Corporation. We are thankful to the anonymous reviewers who gave us the motivation to improve this paper.

Author information

Authors and Affiliations

  1. National Institute of Technology Patna, Patna, India

    Anant Saraswat & Kumar Abhishek

  2. Indian Institute of Technology Kharagpur, Kharagpur, India

    Manjish Pal & Subham Pokhriyal

Authors
  1. Anant Saraswat
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Manjish Pal
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Subham Pokhriyal
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Kumar Abhishek
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Anant Saraswat.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Saraswat, A., Pal, M., Pokhriyal, S. et al. Towards fair machine learning using combinatorial methods. Evol. Intel. 16, 903–916 (2023). https://doi.org/10.1007/s12065-022-00702-5

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12065-022-00702-5

Keywords

Search

Navigation