Research : Machine Learning

Working Papers (and Conference Workshop Presentations)

The Algorithmic Automation Problem: Prediction, Triage, and Human Effort,” with Maithra Raghu, Katy Blumer, Greg Corrado, Jon Kleinberg, and Ziad Obermeyer, 2019.

A Probabilistic Model of Cardiac Physiology and Electrocardiograms,” joint with Andrew Miller, Ziad Obermeyer, David M. Blei, and John P. Cunnigham, arXiv preprint arXiv:1812.00209 (2018).

Measuring the Stability of EHR- and EKG-based Predictive Models,” joint with Andrew C. Miller, Ziad Obermeyer, Sendhil Mullainathan Machine Learning for Health (NeurIPS Workshop), 2018

Simplicity Creates Inequity: Implications for Fairness, Stereotypes, and Interpretability,” with Jon Kleinberg, draft, 2018.

Machine Learning Tests for Effects on Multiple Outcomes,” with Jens Ludwig and Jann Spiess, working paper

A Machine Learning Approach to Low-Value Health Care: Wasted Tests, Missed Heart Attacks and Mis-Predictions,” with Ziad Obermeyer, 2019, NBER working paper 26168, National Bureau of Economics Research, Inc.

Making Sense of Recommendations,” with Mike Yeomans, Anuj Shah and Jon Kleinberg, draft 2016.

The Theory is Predictive, but is it Complete? An Application to Human Perception of Randomness,” with Jon Kleinberg and Annie Liang, draft, 2015.

Publications and Refereed Conference Presentations

The Inversion Problem: Why Algorithms Should Infer Mental State and Not Just Predict Behavior,” with Jon Kleinberg, Jens Ludwig, and Manish Raghavan. Perspectives on Psychological Science, December 2023.

An algorithmic approach to reducing unexplained pain disparities in underserved populations,” with Emma Pierson, David M. Cutler, Jure Leskovec and Ziad Obermeyer, in Nature Medicine volume 27, pages 136–140 (2021)

Allocation of COVID-19 Relief Funding to Disproportionately Black Counties”, with Pragya Kakani, Amitabh Chandra and Ziad Obermeyer. Journal of American Medical Association, August 2020.

Algorithms as discrimination detectors.” with Jon Kleinberg, Jens Ludwig, and Cass R. Sunstein. Proceedings of the National Academy of Sciences July 28 (2020).

An Economic Perspective on Algorithmic Fairness“. with Ashesh Rambachan, Jon Kleinberg and Jens Ludwig, in AEA Papers and Proceedings (Vol. 110, pp. 91-95). [nonrefereed]

Dissecting Racial Bias in an Algorithm Used to Manage the Health of Populations,” with Ziad Obermeyer, Brian Powers, and Christine Vogeli, Science, 366(6464), pp.447-453, 2019.

Direct Uncertainty Prediction for Medical Second Opinions,” with Maithra Raghu, Katy Blumer, Jon Kleinberg, Rory Sayres, Ziad Obermeyer, and Robert Kleinberg, International Conference on Machine Learning (ICML), 2019.

Discriminative Regularization for Latent Variable Models with Applications to Electrocardiography,” with Andrew Miller, Ziad Obermeyer and John Cunningham, International Conference on Machine Learning (ICML), 2019.

Discrimination in the Age of Algorithms,” joint with Jon Kleinberg, Jens Ludwig, and Cass Sunstein, Journal of Legal Studies, 2018.

A Comparison of Patient History- and EKG-based Cardiac Risk Scores,” joint with Andrew C. Miller and Ziad Obermeyer, Proceedings of the AMIA Summit on Clinical Research Informatics (CRI), 2019

Predictive Modeling of US Healthcare Spending in Late Life,” with Liran Einav, Amy Finkelstein and Ziad Obermeyer (2018), Science 29 Jun 2018: Vol. 360, Issue 6396, pp. 1462-1465.  DOI: 10.1126/science.aar5045

Algorithmic Fairness,” joint with Jon Kleinberg, Jens Ludwig and Ashesh Rambachan  (2018), AEA Papers and Proceedings (Vol. 108, pp. 22-27). (Non refereed).

Human Decisions and Machine Predictions,” with Jon Kleinberg, Himabindu Lakkaraju, Jure Leskovec and Jens Ludwig, Quarterly Journal of Economics, 133.1 (2018): 237-293. (also: NBER Working Papers 23180, National Bureau of Economic Research, Inc.)

The Selective Labels Problem: Evaluating Algorithmic Predictions in the Presence of Unobservables,” with Jon Kleinberg, Himabindu Lakkaraju, Jure Leskovec and Jens Ludwig (2017), Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.

Does Machine Learning Automate Moral Hazard and Error?” with Ziad Obermeyer (2017), American Economic Review, May 207, 107 (5): 476-80.  (Non-refereed)

Machine Learning: An Applied Econometric Approach,” with Jann Spiess (2017), Journal of Economic Perspectives, Spring 2017, 31 (2): 87-106.

Inherent Trade-Offs in the Fair Determination of Risk Scores,” with Jon Kleinberg and Manish Raghavan (2017). Proceedings of the 8th Conference on Innovations in Theoretical Computer Science (ITCS).

Assessing Human Error Against a Benchmark of Perfection,” with Ashton Anderson and Jon Kleinberg, Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 2016.

Productivity and Selection of Human Capital with Machine Learning,” with Aaron Chalfin, Oren Danieli, Andrew Hillis, Zubin Jelveh, Michael Luca, and Jens Ludwig, American Economic Review. May 2016, 106 (5): 435-440. (Non refereed)

Prediction Policy Problems,” with Jon Kleinberg, Jens Ludwig and Ziad Obermeyer, American Economic Review (May), 2015.  (Non-refereed)

Other Articles (and More General Writings)

A Guide to Solving Social Problems with Machine Learning” with Ludwig, Jens and Jon Kleinberg. Harvard Business Review December 8 (2016).

Algorithms Need Managers, Too,” with Luca, Michael and Jon Kleinberg. Harvard Business Review 94 (2016): 96-101.

We Built Them, But We Don’t Understand Them,” with Jon Kleinberg, in What to Think About Machines That Think: Today’s Leading Thinkers on the Age of Machine Intelligence. (J. Brockman, editor). Harper Perennial 2015.