Amanda Coston is a PhD student in Machine Learning and Public Policy at Carnegie Mellon University (CMU). Her research considers the impact of algorithmic risk assessments in settings such as child welfare screening, criminal justice, and loan approvals. She is particularly interested in how techniques from causal inference and transfer learning can resolve current limitations of these systems. She is advised by Alexandra Chouldechova and Edward H. Kennedy.
Featured ResearchA. Coston, N. Guha, L. Lu, D. Ouyang, A. Chouldechova, and D. Ho. "Leveraging Administrative Data for Bias Audits: Assessing Disparate Coverage with Mobility Data for COVID-19 Policy." ACM Conference on Fairness, Accountability, and Transparency, 2021. [Paper] [ArXiv] [Talk]
A. Coston, A. Rambachan, and A. Chouldechova. "Characterizing Fairness over the Set of Good Models Under Selective Labels." International Conference on Machine Learning, 2021 (to appear). [ArXiv]
A. Coston, E. H. Kennedy, and A. Chouldechova. "Counterfactual Predictions under Runtime Confounding." Neural Information Processing Systems, 2020. [Paper] [ArXiv] [Blog]
A. Coston, A. Mishler, E. H. Kennedy, and A. Chouldechova. "Counterfactual Risk Assessments, Evaluation, and Fairness." ACM FAT* 2020. [Paper] [ArXiv] [Talk]
A. Coston, K. N. Ramamurthy, D. Wei, K. R. Varshney, S. Speakman, Z. Mustahsan, S. Chakraborty. "Fair Transfer Learning with Missing Protected Attributes," AAAI/ACM Conference on Artificial Intellligence, Ethics, and Society (AIES), 2019. [Paper]
- June 7, 2021 Starting internship at Facebook Responsible AI
- May 18, 2021 Featured on Placekey Spotlight
- May 8, 2021 Research on characterizing fairness over the set of good models under selective labels accepted at ICML
- May 4, 2021 Invited talk at Johns Hopkins Causal Inference Working Group on counterfactual predictions for decision-making [Video]
- April 22, 2021 Invited talk at PlaceKey COVID-19 Research Consortium on auditing mobility data for disparate coverage by race and age [Video]
- April 16, 2021 CMU ML blog post on counterfactual predictions under runtime confounding
- April 5, 2021 Wall Street Journal piece featured her research on auditing mobility data for demographic bias "Smartphone Location Data Can Leave Out Those Most Hit by Covid-19"
- November 18, 2020 VentureBeat piece featured her research on auditing mobility data for demographic bias “Stanford and Carnegie Mellon find race and age bias in mobility data that drives COVID-19 policy”