Manuscripts not linked to are available on request.
Algorithmic Fairness
An Impossibility Theorem for Base Rate Tracking and Equalised Odds (with Ben Eva, Shanna Slank, Reuben Stern) [Preprint] [Journal]
Identity and the Limits of Fair Assessment [Preprint] [Journal]
New Possibilities for Fair Algorithms (with Michael Nielsen) [Preprint] [Journal]
On the Possibility of Testimonial Justice (with Michael Nielsen) [Preprint] [Journal]
Spanning in and Spacing out? A Reply to Eva (with Michael Nielsen) [Preprint] [Journal]
The Ideals Program in Algorithmic Fairness [Preprint] [Journal]
Decision Theory
Probability
Conglomerability, Disintegrability, and the Comparative Principle (with Michael Nielsen) [Preprint] [Journal]
Counterexamples to Some Characterizations of Dilation (with Michael Nielsen) [Preprint] [Journal]
Distention for Sets of Probabilities (with Michael Nielsen) [Preprint] [Journal]
Obligation, Permission, and Bayesian Orgulity (with Michael Nielsen) [Preprint] [Journal]
Peirce, Pedigree, Probability (with Tom F. Sterkenburg) [Preprint] [Journal]
Social Epistemology
Another Approach to Consensus and Maximally Informed Opinions with Increasing Evidence (with Michael Nielsen) [Preprint] [Journal]
Learning and Pooling, Pooling and Learning (with Ignacio Ojea Quintana) [Preprint] [Journal]
Persistent Disagreement and Polarization in a Bayesian Setting (with Michael Nielsen) [Preprint] [Journal]
Probabilistic Opinion Pooling with Imprecise Probabilities (with Ignacio Ojea Quintana) [Preprint] [Journal]
Support for Geometric Pooling (with Jean Baccelli) [Preprint] [Journal]
Unanimous Consensus against AGM? [Preprint] [Journal]
What’s Hot in Mathematical Philosophy Column in The Reasoner 12(10) (with Michael Nielsen) [Gazette]
Dissertation
Mathematical aggregation frameworks are general and precise settings in which to study ways of forming a consensus or group point of view from a set of potentially diverse points of view. Yet the standard frameworks have significant limitations. A number of results show that certain sets of desirable aggregation properties cannot be simultaneously satisfied. Drawing on work in the theory of imprecise probabilities, I propose philosophically-motivated generalizations of the standard aggregation frameworks (for probability, preference, full belief) that I prove can satisfy the desired properties. I then look at some applications and consequences of these proposals in decision theory, epistemology, and the social sciences. [Short Abstract] [Extended Abstract]