Padhraic Smyth
Chancellor's Professor
Department of Computer Science
Department of Statistics
School of Information and Computer Sciences
University of California, Irvine
- Associate Director, Center for Machine Learning and Intelligent Systems
- Affiliate Faculty, UCI Steckler Center for Responsible, Ethical, and Accessible Technology (CREATe)
- Member, Institute for Mathematical Behavioral Sciences at UCI
- Our group is part of a new National Institutes of Health (NIH) 4-year research project on machine learning and fairness in predictive risk modeling for health outcomes [Fall 2021]
- Looking forward to participating in exciting new projects as part of the newly-founded CREATe Center at UCI on social aspects of AI and machine learning [Winter 2021]
- Our research group is part of the UCI team publishing daily statistics to increase public awareness about the the latest COVID-19 trends in Southern California - great work by various PhD students in statistics and computer science and Prof Vladimir Minin [Summer/Fall 2020]
- The AI/ML group at UC Irvine is part of the new HPI Research Center in Machine Learning and Data Science, providing 15 fully-funded 3-year fellowships for UCI PhD students in machine learning and databases [Spring 2020]
- Grateful for gift funding from Qualcomm to support our group's research via a Qualcomm faculty award [2019-2021]
Selected Recent Papers
-
Bayesian modeling of human-AI complementarity
with Mark Steyvers, Heliodoro Tejeda, and Gavin Kerrigan, Proceedings of the National Academy of Sciences, 2022 -
Combining human predictions with model probabilities via confusion matrices and calibration
with Gavin Kerrigan and Mark Steyvers, Advances in Neural Information Processing Systems 34 (NeurIPS), 2021 - Active Bayesian assessment of black-box classifiers
with Disi Ji, Robby Logan, and Mark Steyvers, 35th AAAI Conference on Artificial Intelligence (AAAI), 2021 -
Can I trust my fairness metric? Assessing fairness with unlabeled data and Bayesian inference
with Disi Ji and Mark Steyvers, Advances in Neural Information Processing Systems 33 (NeurIPS), 2020 -
User-dependent neural sequence models for continuous-time event data
with Alex Boyd, Rob Bamler, Stephan Mandt, Advances in Neural Information Processing Systems 33 (NeurIPS), 2020 - Dropout as a structured shrinkage prior
with Eric Nalisnick and Jose Miguel Hernandez-Lobato, International Conference on Machine Learning (ICML), 2019
Teaching