Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
12 tokens/sec
GPT-4o
12 tokens/sec
Gemini 2.5 Pro Pro
41 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
37 tokens/sec
DeepSeek R1 via Azure Pro
33 tokens/sec
2000 character limit reached

Toward A Logical Theory Of Fairness and Bias (2306.13659v1)

Published 8 Jun 2023 in cs.AI and cs.LO

Abstract: Fairness in machine learning is of considerable interest in recent years owing to the propensity of algorithms trained on historical data to amplify and perpetuate historical biases. In this paper, we argue for a formal reconstruction of fairness definitions, not so much to replace existing definitions but to ground their application in an epistemic setting and allow for rich environmental modelling. Consequently we look into three notions: fairness through unawareness, demographic parity and counterfactual fairness, and formalise these in the epistemic situation calculus.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (64)
  1. Deontological ethics. In The Stanford Encyclopedia of Philosophy. 2016.
  2. Geneth: A general ethical dilemma analyzer. In Proceedings of the 28th AAAI Conference on Artificial Intelligence, pages 253–261, 2014.
  3. “machine bias: There’s software used across the country to predict future criminals. and it’s biased against blacks.”. 2016.
  4. Knowledge, fairness, and social constraints. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 32, 2018.
  5. Reasoning about noisy sensors and effectors in the situation calculus. Artificial Intelligence, 111(1–2):171 – 208, 1999.
  6. From statistical knowledge bases to degrees of belief. Artificial intelligence, 87(1-2):75–143, 1996.
  7. Exploiting procedural domain control knowledge in state-of-the-art planners. In Proc. ICAPS, pages 26–33, 2007. URL http://www.aaai.org/Library/ICAPS/2007/icaps07-004.php.
  8. Epistemic Planning (Dagstuhl Seminar 17231). Dagstuhl Reports, 7(6):1–47, 2017. ISSN 2192-5283. 10.4230/DagRep.7.6.1. URL http://drops.dagstuhl.de/opus/volltexte/2017/8285.
  9. Situation calculus semantics for actual causality. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 32, 2018.
  10. V. Belle and G. Lakemeyer. Reasoning about probabilities in unbounded first-order dynamical domains. In IJCAI, 2017.
  11. Vaishak Belle. Analyzing generalized planning under nondeterminism. Artificial Intelligence, page 103696, 2022.
  12. The boid architecture: conflicts between beliefs, obligations, intentions and desires. In Proceedings of the fifth international conference on Autonomous agents, pages 9–16, 2001.
  13. Responsibility and blame: A structural-model approach. Journal of Artificial Intelligence Research, 22:93–115, 2004.
  14. A. Chouldechova. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. In Big Data, Volume 5, No. 2: 153–163, 2017.
  15. Jens Classen. Symbolic verification of Golog programs with first-order BDDs. In Michael Thielscher, Francesca Toni, and Frank Wolter, editors, Proceedings of the Sixteenth International Conference on Principles of Knowledge Representation and Reasoning (KR 2018), pages 524–528. AAAI Press, 2018. URL http://jens-classen.net/pub/Classen2018.pdf.
  16. Dyadic obligations over complex actions as deontic constraints in the situation calculus. In Proceedings of the International Conference on Principles of Knowledge Representation and Reasoning, volume 17, pages 253–263, 2020.
  17. A logic for non-terminating golog programs. In KR, pages 589–599, 2008.
  18. Integrating Golog and planning: An empirical evaluation. In NMR Workshop, pages 10–18, 2008.
  19. Deontological and utilitarian inclinations in moral decision making: a process dissociation approach. Journal of Personality and Social Psychology, 104(2):216, 2013.
  20. Causal modeling for fairness in dynamical systems. In International Conference on Machine Learning, pages 2185–2195. PMLR, 2020.
  21. Janusz Czelakowski. Action and deontology. In Logic, action and cognition, pages 47–87. Springer, 1997.
  22. An integrated reasoning approach to moral decision-making. In Proceedings of the 23rd AAAI Conference on Artificial Intelligence, pages 1280–1286, 2008.
  23. Belief revision with sensing and fallible actions. In Proc. KR, 2012.
  24. Fairness through awareness. In Innovations in Theoretical Computer Science Conference, 2011.
  25. Fairness in relational domains. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, pages 108–114, 2018.
  26. Fairness-aware explainable recommendation over knowledge graphs. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 69–78, 2020.
  27. The belief-desire-intention model of agency. In International workshop on agent theories, architectures, and languages, pages 1–10. Springer, 1998.
  28. J. Y. Halpern and Y. Moses. A procedural characterization of solution concepts in games. JAIR, 49:143–170, 2014.
  29. An epistemic characterization of zero knowledge. In TARK, pages 156–165, 2009.
  30. Towards formal definitions of blameworthiness, intention, and moral responsibility. In Proceedings of the 32nd AAAI Conference on Artificial Intelligence, pages 1853–1860, 2018.
  31. Equality of opportunity in supervised learning. In International Conference on Neural Information Processing Systems, 2016.
  32. John N Hooker and Tae Wan N Kim. Toward non-intuition-based machine and artificial intelligence ethics: A deontological approach based on modal logic. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, pages 130–136, 2018.
  33. Towards formal fairness in machine learning. In Principles and Practice of Constraint Programming: 26th International Conference, CP 2020, Louvain-la-Neuve, Belgium, September 7–11, 2020, Proceedings 26, pages 846–867. Springer, 2020.
  34. Yusuke Kawamoto. Towards logical specification of statistical machine learning. In Software Engineering and Formal Methods: 17th International Conference, SEFM 2019, Oslo, Norway, September 18–20, 2019, Proceedings, pages 293–311. Springer, 2019.
  35. Consumer credit-risk models via machine-learning algorithms. In Journal of Banking and Finance 34, 2010.
  36. Counterfactual fairness. Neural Information Processing Systems, 2017.
  37. G. Lakemeyer and H. J. Levesque. Cognitive robotics. In Handbook of Knowledge Representation, pages 869–886. Elsevier, 2007.
  38. G. Lakemeyer and H. J. Levesque. A semantic characterization of a useful fragment of the situation calculus with knowledge. Artificial Intelligence, 175:142–164, 2011. ISSN 0004-3702.
  39. G. Lakemeyer and Hector J. Levesque. Situations, Si! situation terms, No! In Proc. KR, pages 516–526, 2004.
  40. Reformulating the situation calculus and the event calculus in the general theory of stable models and in answer set programming. Journal of Artificial Intelligence Research, 43:571–620, 2012.
  41. Ability and knowing how in the situation calculus. Studia Logica, 66(1):165–186, 2000.
  42. Golog: A logic programming language for dynamic domains. Journal of Logic Programming, 31:59–84, 1997. URL GOLOGlang.pdf.
  43. H. J. Levesque. All I know: a study in autoepistemic logic. Artificial Intelligence, 42(2-3):263–309, 1990. ISSN 0004-3702.
  44. H. J. Levesque. What is planning in the presence of sensing? In Proc. AAAI / IAAI, pages 1139–1146, 1996.
  45. F. Lin and R. Reiter. Forget it. In Working Notes of AAAI Fall Symposium on Relevance, pages 154–159, 1994.
  46. The hera approach to morally competent robots. In Proceedings of the IEEE/RSJ Intelligent Robots and Systems, pages 6991–6997, 2017.
  47. A logic of “black box” classifier systems. In Logic, Language, Information, and Computation: 28th International Workshop, WoLLIC 2022, Iași, Romania, September 20–23, 2022, Proceedings, pages 158–174. Springer, 2022.
  48. Adapting golog for composition of semantic web services. In KR, pages 482–496, 2002.
  49. Ilp turns 20. Machine learning, 86(1):3–23, 2012.
  50. Planning over multi-agent epistemic states: A classical planning approach. In Proc. AAAI, 2015.
  51. Epistemic reasoning for machine ethics with situation calculus. In Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society, pages 814–821, 2021.
  52. Discrimination aware data mining. In 14th ACM SIGKDD international conference on Knowledge Discovery and Data Mining, 2008.
  53. Statistical relational artificial intelligence: Logic, probability, and computation. Synthesis lectures on artificial intelligence and machine learning, 10(2):1–189, 2016.
  54. R. Reiter. On knowledge-based programming with sensing in the situation calculus. ACM Trans. Comput. Log., 2(4):433–457, 2001a.
  55. R. Reiter. Knowledge in action: logical foundations for specifying and implementing dynamical systems. MIT Press, 2001b.
  56. On the semantics of deliberation in indigolog—from theory to implementation. Annals of Mathematics and Artificial Intelligence, 41(2-4):259–299, 2004.
  57. Golog speaks the bdi language. In Programming Multi-Agent Systems, volume 5919 of LNCS, pages 82–99. Springer Berlin Heidelberg, 2010. ISBN 978-3-642-14842-2.
  58. Knowledge, action, and the frame problem. Artificial Intelligence, 144(1-2):1–39, 2003. ISSN 0004-3702.
  59. Fairness definitions explained. In 2018 ieee/acm international workshop on software fairness (fairware), pages 1–7. IEEE, 2018.
  60. Nested epistemic logic programs. In International Conference on Logic Programming and Nonmonotonic Reasoning, pages 279–290. Springer, 2005.
  61. Fairness beyond disparate treatment & disparate impact: Learning classification without disparate mistreatment. In International Conference on World Wide Web, 2017a.
  62. International conference on artificial intelligence and statistics. In Fairness Constraints: Mechanisms for Fair Classification, 2017b.
  63. Learning fair representations. In International Conference on Machine Learning, 2013.
  64. Yan Zhang and Yi Zhou. Knowledge forgetting: Properties and applications. Artificial Intelligence, 173(16-17):1525–1537, 2009.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now