Fairness and reliability of Machine Learning

Fairness criteria: Independence, separation, sufficiency. Impossibility. (The Fair ML Book (Chapter 3))
Limitations of observational criteria. (The Fair ML Book (Chapter 3), NIPS 2017 tutorial, also see The Book of Why: The New Science of Cause and Effect by Judea Pearl and Dana Mackenzie.)
Causal inference, counterfactual reasoning. (An intro to causal inference by J Pearl (also see the book on Causality), The Fair ML Book (Chapter 4))
Discussions on instrumental variables, regression discontinuity and randomzed controlled trials (RCTs) .
Fairness through awareness by Dwork et al.