⚖ Fairness Journal Club

Co-orgainzed with Shems Saleh and David Madras .

We read papers related to algorithmic bias and the design of "fair" machine learning algorithms. Here's a list of what we've covered thusfar:

• On October 23, 2018 we discussed Fairness Without Demographics in Repeated Loss Minimization by Hashimoto et al. (Stanford).
• On October 2, 2018 we discussed Causal Interventions for Fairness by Kusner et al. (Alan Turing Institute, U. Warwick, NYU, U. Surrey, UCL).
• On September 11 & 18, 2018 we discussed Multiaccuracy: Black-Box Post-Processing for Fairness in Classification by Kim et al. (Stanford).
• On April 10, 2018 we discussed Delayed Impact of Fair Machine Learning by Liu et al. (UC Berkeley).
• On March 20 & 27, 2018 we discussed Certifying and removing disparate impact by Feldman et al. (Haverford, U. Utah, U. Arizona).
• On March 13, 2018 we discussed Indirect Discrimination and the Duty to Avoid Compounding Injustice by Hellman (U. Virginia).
• On March 6, 2018 we discussed Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings by Bolukbasi et al. (Boston U., Microsoft).
• On February 27, 2018 David and Elliot provided a recap of the papers covered so far.
• On February 20, 2018 we discussed Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness by Kearns et al. (U. Pennslyvania, Microsoft).
• On February 13, 2018 we discussed Counterfactual Fairness by Kusner et al. (Alan Turing Institute, U. Warwick, NYU, U. Surrey, UCL).
• On January 30, 2018 we discussed Equality of Opportunity in Supervised Learning by Hardt et al. (Google, UT Austin, TTI-Chicago).
• On January 23, 2018 we discussed Fairness Through Awareness by Dwork et al. (Microsoft, IBM, U. Toronto).
• On January 16, 2018 we discussed Fair prediction with disparate impact: A study of bias in recidivism prediction instruments by Chouldechova (CMU).