Golnoosh Farnadi
ABSTRACT
AI and machine learning tools are being used with increasing frequency for decision making in domains that affect peoples' lives such as employment, education, policing and loan approval. These uses raise concerns about biases of algorithmic discrimination and have motivated the development of fairness-aware machine learning. However, existing fairness approaches are based solely on attributes of individuals. In many cases, discrimination is much more complex, and taking into account the social, organizational, and other connections between individuals is important. We introduce new notions of fairness that are able to capture the relational structure in a domain. We use first-order logic to provide a flexible and expressive language for specifying complex relational patterns of discrimination. Furthermore, we extend an existing statistical relational learning framework, probabilistic soft logic (PSL), to incorporate our definition of relational fairness. We refer to this fairness-aware framework FairPSL. FairPSL makes use of the logical definitions of fairnesss but also supports a probabilistic interpretation. In particular, we show how to perform maximum a posteriori(MAP) inference by exploiting probabilistic dependencies within the domain while avoiding violation of fairness guarantees. Preliminary empirical evaluation shows that we are able to make both accurate and fair decisions.
- European Union Legislation. (a) Racial Equality Directive. 2000. (b) Employment Equality Directive, 2000. (c) Gender Employment Directive, 2006. (d) Equal Treatment Directive (proposal), 2008.Google Scholar
- UK Legislation. (a) Sex Discrimination Act. 1975. (b) Race Relation Act, 1976.Google Scholar
- United Nations Legislation. (a) Universal Declaration of Human Rights. 1948. (c) Convention on the Elimination of All forms of Racial Discrimination, 1966. (d) Convention on the Elimination of All forms of Discrimination Against Women, 1979.Google Scholar
- Duhai Alshukaili, Alvaro A. A. Fernandes, and Norman W. Paton. 2016. Structuring Linked Data Search Results Using Probabilistic Soft Logic. In International Semantic Web Conference (1) (Lecture Notes in Computer Science), Vol. 9981. 3--19.Google Scholar
- Stephen H. Bach, Matthias Broecheler, Bert Huang, and Lise Getoor. 2017. Hinge- Loss Markov Random Fields and Probabilistic Soft Logic. Journal of Machine Learning Research 18 (2017), 109:1--109:67. Google ScholarDigital Library
- Solon Barocas and Andrew D Selbst. 2016. Big data's disparate impact. California Law Review 104 (2016), 671.Google Scholar
- Danah Boyd, Karen Levy, and Alice Marwick. 2014. The networked nature of algorithmic discrimination. In Data and discrimination: Collected essays. 53--57.Google Scholar
- YooJung Choi, Adnan Darwiche, and Guy Van den Broeck. 2017. Optimal Feature Selection for Decision Robustness in Bayesian Networks. In IJCAI. ijcai.org, 1554--1560. Google ScholarDigital Library
- Alexandra Chouldechova. 2017. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. CoRR abs/1703.00056 (2017).Google Scholar
- Steven Diamond and Stephen P. Boyd. 2016. CVXPY: A Python-Embedded Modeling Language for Convex Optimization. Journal of Machine Learning Research 17 (2016), 83:1--83:5. Google ScholarDigital Library
- Cynthia Dwork, Moritz Hardt, Toniann Pitassi, Omer Reingold, and Richard S. Zemel. 2012. Fairness through awareness. In ITCS. ACM, 214--226. Google ScholarDigital Library
- Javid Ebrahimi, Dejing Dou, and Daniel Lowd. 2016. Weakly Supervised Tweet Stance Classification by Relational Bootstrapping. In EMNLP. The Association for Computational Linguistics, 1012--1017.Google Scholar
- Michael Feldman, Sorelle A. Friedler, John Moeller, Carlos Scheidegger, and Suresh Venkatasubramanian. 2015. Certifying and Removing Disparate Impact. In KDD. ACM, 259--268. Google ScholarDigital Library
- Lise Getoor and Ben Taskar. 2007. Introduction to Statistical Relational Learning. Vol. 1. MIT press Cambridge. Google ScholarDigital Library
- Moritz Hardt, Eric Price, and Nati Srebro. 2016. Equality of Opportunity in Supervised Learning. In NIPS. 3315--3323. Google ScholarDigital Library
- Toshihiro Kamishima, Shotaro Akaho, and Jun Sakuma. 2011. Fairness-aware Learning through Regularization Approach. In ICDMW. IEEE Computer Society, 643--650. Google ScholarDigital Library
- Pigi Kouki, Shobeir Fakhraei, James R. Foulds, Magdalini Eirinaki, and Lise Getoor. 2015. HyPER: A Flexible and Extensible Probabilistic Framework for Hybrid Recommender Systems. In RecSys. ACM, 99--106. Google ScholarDigital Library
- Matt J. Kusner, Joshua R. Loftus, Chris Russell, and Ricardo Silva. 2017. Counterfactual Fairness. In NIPS. 4069--4079. Google ScholarDigital Library
- Dino Pedreschi, Salvatore Ruggieri, and Franco Turini. 2012. A study of top-k measures for discrimination discovery. In SAC. ACM, 126--131. Google ScholarDigital Library
- Dino Pedreschi, Salvatore Ruggieri, and Franco Turini. 2013. The Discovery of Discrimination. In Discrimination and Privacy in the Information Society. Studies in Applied Philosophy, Epistemology and Rational Ethics, Vol. 3. Springer, 91--108.Google Scholar
- Dhanya Sridhar, Shobeir Fakhraei, and Lise Getoor. 2016. A probabilistic approach for collective similarity-based drug-drug interaction prediction. Bioinformatics 32, 20 (2016), 3175--3182.Google ScholarCross Ref
- RobertWest, Hristo S. Paskov, Jure Leskovec, and Christopher Potts. 2014. Exploiting Social Network Structure for Person-to-Person Sentiment Analysis. TACL 2 (2014), 297--310.Google ScholarCross Ref
- Muhammad Bilal Zafar, Isabel Valera, Manuel Gomez-Rodriguez, Krishna P. Gummadi, and Adrian Weller. 2017. From Parity to Preference-based Notions of Fairness in Classification. In NIPS. 228--238. Google ScholarDigital Library
- Richard S. Zemel, Yu Wu, Kevin Swersky, Toniann Pitassi, and Cynthia Dwork. 2013. Learning Fair Representations. In ICML (3) (JMLRWorkshop and Conference Proceedings), Vol. 28. JMLR.org, 325--333. Google ScholarDigital Library
Recommendations
Airtime Fairness for IEEE 802.11 Multirate Networks
Under a multi rate network scenario, the IEEE 802.11 DCF MAC fails to provide air-time fairness for all competing stations since the protocol is designed for ensuring max-min throughput fairness and the maximum achievable throughput by any station gets ...
Statistical Relational Learning: Unifying AI & DB Perspectives on Structured Probabilistic Models
PODS '17: Proceedings of the 36th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database SystemsMachine learning and database approaches to structured probabilistic models share many commonalities, yet exhibit certain important differences. Machine learning methods focus on learning probabilistic models from (certain) data and efficient learning ...
Fairness in multi-hop wireless backhaul networks: a dynamic estimation approach
QShine '08: Proceedings of the 5th International ICST Conference on Heterogeneous Networking for Quality, Reliability, Security and RobustnessIn this work, we consider the problem of fairness for Transit Access Points (TAP) in multi-hop wireless backhaul networks. Existing approaches are not practical due to the requirement for modifications to the MAC layer or queueing operations of TAPs, or ...
Comments