Padhraic Smyth
Distinguished Professor of Computer Science and Director of the Data Science Initiative UC Irvine
- Irvine CA
Padhraic Smyth studies machine learning, artificial intelligence, pattern recognition and applied statistics.
Media
Social
Biography
Padhraic was the founding director of the UCI Center for Machine Learning and Intelligent Systems from 2007 to 2014 and founding director from 2014 to 2018 of the UCI Data Science Initiative. While at UC Irvine he has received research funding from agencies such as NSF, NIH, IARPA, NASA, NIST, ONR, and DOE, and from companies such as Google, Qualcomm, Microsoft, eBay, Adobe, IBM, SAP, Xerox, and Experian. In addition to his academic research he is also active in industry consulting, working with companies such as Toshiba, Samsung, Oracle, Nokia, and AT&T, as well as serving as scientific advisor to local startups in Orange County. He also served as an academic advisor to Netflix for the Netflix prize competition from 2006 to 2009.
Padhraic grew up in Kilmovee, County Mayo, in the west of Ireland and received a first class honors degree in Electronic Engineering from National University of Ireland (University of Galway) in 1984, and the MSEE and PhD degrees (in 1985 and 1988 respectively) in Electrical Engineering from the California Institute of Technology. From 1988 to 1996 he was a researcher at the Jet Propulsion Laboratory, Pasadena, and has been on the faculty at UC Irvine since 1996.
Areas of Expertise
Accomplishments
Outstanding Paper with Lead Student Author, International Conference on AI and Statistics
2024
Qualcomm Faculty Award
2019, 2020, 2021
Fellow, Institute for Electrical and Electronic Engineers (IEEE)
2023
Fellow, American Association for the Advancement of Science (AAAS)
2022
Best Paper, Educational Data Mining Conference (EDM)
2018
Education
California Institute of Technology
Ph.D.
Electrical Engineering
1988
California Institute of Technology
MSEE
Electrical Engineering
1985
National University of Ireland
BE
Engineering (Electronic)
1984
Affiliations
- AAAI President’s Fellows Advisory Board : Member, 2019–present
Media Appearances
How UCI and AI go waaay back
Newswise online
2023-12-04
“A lot of the ideas we’re using now were around then,” says Padhraic Smyth, Chancellor’s Professor of computer science. “But they couldn’t do them in the 1960s because computers were too primitive.”
A call to hit the ‘pause’ button on AI experiments
University of California online
2023-04-07
Chancellor’s professor of computer science Padhraic Smyth, who also signed the open letter, agrees.
“My concern is that we are far from a full understanding of the limitations and dangers of these models,” says Smyth. “As AI researchers, we understand how to write down the mathematics and algorithms so that these models can be learned from data, but once they are learned, we don’t have the tools to understand and characterize what they are capable of,” he says. “Now is the time to take a closer look at the issue of AI safety and put the common good ahead of commercial interests.”
Event Appearances
Bayesian consensus prediction for correlated human experts and classifiers
2025 | International Conference on Machine Learning (ICML) Vancouver, Canada
Understanding gender bias in AI-generated product description
2025 | ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT) Athens, Greece
ELBOing Stein: Variational Bayes with Stein mixture inference
2025 | International Conference on Learning Representations Singapore
Dynamic conditional optimal transport through simulation-free flows
2024 | Proceedings of the Thirty-Eighth Annual Conference on Neural Information Processing Systems Vancouver, Canada
Benchmark data repositories for better benchmarking
2024 | Proceedings of the Thirty-Eighth Annual Conference on Neural Information Processing Systems Vancouver, Canada
Patents
Hidden Markov Models for Fault Detection in Dynamic Systems,
U.S. Patent no. 5465321
November 7 1995
Cross-Connect Switch and Method for Providing Test Access Thereto,
U.S. Patent no. 4845736
Issued July 4 1989
Cross-Connect Switch
U.S. Patent no. 4807280
Issued February 21 1989
Research Grants
CAIG: Advancing Wildfire Science, Prediction, and Management with Machine Learning
NSF 2425932
10/2024-9/2027
Time Series Prediction with Deep Learning
9/2024
Individualized Learning and Prediction for Heterogeneous Multimodal Data from Wearable Devices
NIH R01 CA297869-01,
7/2024-6/2028
AI/ML and Data Science Training Datasets
NIH 3OT2OD032581,
1/2023-3/2024
Improving Prediction of Fire Extremes in the GEOS Forecasting System on Daily and Seasonal Timescales,
NASA
9/2021-6/2025
Articles
JANET: Joint adaptive prediction-region estimation for time-series
Machine LearningEshant English, Eliot Wong-Toi, Matteo Fontana, Stephan Mandt, Padhraic Smyth, Christoph Lippert
2025
Conformal prediction provides machine learning models with prediction sets that offer theoretical guarantees, but the underlying assumption of exchangeability limits its applicability to time series data. Furthermore, existing approaches struggle to handle multi-step ahead prediction tasks, where uncertainty estimates across multiple future time points are crucial. We propose JANET (Joint Adaptive predictioN-region Estimation for Time-series), a novel framework for constructing conformal prediction regions that are valid for both univariate and multivariate time series. JANET generalises the inductive conformal framework and efficiently produces joint prediction regions with controlled K-familywise error rates, enabling flexible adaptation to specific application needs. Our empirical evaluation demonstrates JANET's superior performance in multi-step prediction tasks across diverse time series datasets, highlighting its potential for reliable and interpretable uncertainty quantification in sequential data.
What large language models know and what people think they know
Nature Machine IntelligenceMark Steyvers, Heliodoro Tejeda, Aakriti Kumar, Catarina Belem, Sheer Karny, Xinyue Hu, Lukas W Mayer, Padhraic Smyth
2025
As artificial intelligence systems, particularly large language models (LLMs), become increasingly integrated into decision-making processes, the ability to trust their outputs is crucial. To earn human trust, LLMs must be well calibrated such that they can accurately assess and communicate the likelihood of their predictions being correct. Whereas recent work has focused on LLMs’ internal confidence, less is understood about how effectively they convey uncertainty to users. Here we explore the calibration gap, which refers to the difference between human confidence in LLM-generated answers and the models’ actual confidence, and the discrimination gap, which reflects how well humans and models can distinguish between correct and incorrect answers.
A generative diffusion model for probabilistic ensembles of precipitation maps conditioned on multisensor satellite observations
IEEE Transactions on Geoscience and Remote SensingClement Guilloteau, Gavin Kerrigan, Kai Nelson, Giosue Migliorini, Padhraic Smyth, Runze Li, Efi Foufoula-Georgiou
2025
A generative diffusion model is used to produce probabilistic ensembles of precipitation intensity maps at the 1-h 5-km resolution. The generation is conditioned on infrared and microwave radiometric measurements from the GOES and DMSP satellites and is trained with merged ground radar and gauge data over the southeastern United States. The generated precipitation maps reproduce the spatial autocovariance and other multiscale statistical properties of the gauge-radar reference fields on average. Conditioning the generation on the satellite measurements allows us to constrain the magnitude and location of each generated precipitation feature. The mean of the 128-member ensemble shows high spatial coherence with the reference fields with a 0.82 linear correlation between the two.
Dynamic conditional optimal transport through simulation-free flows
Advances in Neural Information Processing SystemsGavin Kerrigan, Giosue Migliorini, Padhraic Smyth
2024
We study the geometry of conditional optimal transport (COT) and prove a dynamic formulation which generalizes the Benamou-Brenier Theorem. Equipped with these tools, we propose a simulation-free flow-based method for conditional generative modeling. Our method couples an arbitrary source distribution to a specified target distribution through a triangular COT plan, and a conditional generative model is obtained by approximating the geodesic path of measures induced by this COT plan. Our theory and methods are applicable in infinite-dimensional settings, making them well suited for a wide class of Bayesian inverse problems. Empirically, we demonstrate that our method is competitive on several challenging conditional generation tasks, including an infinite-dimensional inverse problem.
Likelihood ratios for changepoints in categorical event data with applications in digital forensics
Journal of Forensic SciencesRachel Longjohn, Padhraic Smyth
2024
We investigate likelihood ratio models motivated by digital forensics problems involving time‐stamped user‐generated event data from a device or account. Of specific interest are scenarios where the data may have been generated by a single individual (the device/account owner) or by two different individuals (the device/account owner and someone else), such as instances in which an account was hacked or a device was stolen before being associated with a crime. Existing likelihood ratio methods in this context require that a precise time is specified at which the device or account is purported to have changed hands (the changepoint)—this is the known changepoint likelihood ratio model. In this paper, we develop a likelihood ratio model that instead accommodates uncertainty in the changepoint using Bayesian techniques, that is, an unknown changepoint likelihood ratio model.
