Assistant Professor, Computer Science & Laboratory Medicine and Pathobiology
Canada CIFAR AI Chair at the Vector Institute
Tier II Canada Research Chair in Computational Medicine
I will hire a graduate student in Fall 2026. Apply to the DCS graduate program. See more information on the joining page.
Research Direction
Data collected from natural phenomena characterize computation occurring within complex processes governed by known and unknown laws. Neural networks are powerful tools that can learn to compress computation happening in nature. My group advances fundamental research in machine learning to understand, identify, and create controllable artificial intelligence systems. We are currently focused on building neural network models that reason causally, and apply them to solve problems in healthcare and biology.
Students
PhD Students
- Michael Cooper (co-sup w/ Mike Brudno)
- Vahid Balazadeh-Meresht
- Jerry Ji (co-sup w/ Anna Goldenberg)
- Viet Nguyen
- Lance Chao
- Yujia Ma (co-sup w/ Chris Maddison)
- Steven Palayew (sup by Mike Wainberg)
- Mohammad Adnan (PhD student at the University of Calgary sup. by Yani Ioannou)
Postdoctoral Fellows
MSc Students
Alumni
Former undergraduate students from my lab have gone on to graduate (MSc and PhD) programs (NYU, Princeton, CMU, UC Berkeley, Cambridge, Imperial, MILA, ETH, Harvard, Cornell, Waterloo) and research roles in industry (Cerebras Systems, Google Research, Vanguard, Georgian, Scale AI).
- Tom Ginsberg — MSc (CTO, BeIT Canada)
- Asic Chen — MSc (Google DeepMind)
- Ian Shi — PhD (Startup @ YCombinator)
- Hamidreza Kamkari — MScAC (Layer6; now PhD student at MIT)
- Changjian Shui — Vector Postdoctoral Fellow (now Assistant Professor, University of Ottawa)
Links
Selected Publications
A selected list of representative papers is available below. For a full list, see my Google Scholar profile.
-
CausalPFN: Amortized Causal Effect Estimation via In-Context LearningNeural Information Processing Systems (NeurIPS), 2025(Spotlight)
-
Beyond Masked and Unmasked: Discrete Diffusion Models via Partial Masking.Neural Information Processing Systems (NeurIPS), 2025
-
Reliably Detecting Model Failures in Deployment Without LabelsNeural Information Processing Systems (NeurIPS), 2025
-
Sparse Training from Random Initialization: Aligning Lottery Ticket Masks using Weight SymmetryInternational Conference on Machine Learning (ICML) 2025
-
AutoElicit: Using Large Language Models for Expert Prior Elicitation in Predictive ModellingInternational Conference on Machine Learning (ICML) 2025
-
Diverse Prototypical Ensembles Improve Robustness to Subpopulation ShiftInternational Conference on Machine Learning (ICML) 2025
-
ExOSITO: Explainable Off-Policy Learning with Side Information for Intensive Care Unit Blood Test OrdersConference on Health, Inference and Learning (CHIL), 2025
-
Teaching LLMs How to Learn with Contextual Fine-TuningInternational Conference on Learning Representations (ICLR), 2025
-
End-To-End Causal Effect Estimation from Unstructured Natural Language DataNeural Information Processing Systems (NeurIPS), 2024
-
Sequential Decision Making with Expert Demonstrations under Unobserved HeterogeneityNeural Information Processing Systems (NeurIPS), 2024 ( *: equal contribution )
-
Long-Term Allograft Survival in Liver Transplant RecipientsMachine Learning for Healthcare (MLHC), 2024
-
NeRF-US: Removing Ultrasound Imaging Artifacts from Neural Radiance Fields in the WildMachine Learning for Healthcare (MLHC), 2024
-
InterpreTabNet: Distilling Predictive Signals from Tabular Data by Salient Feature InterpretationInternational Conference on Machine Learning (ICML), 2024(Spotlight)
-
A Geometric Explanation of the Likelihood OOD Detection ParadoxInternational Conference on Machine Learning (ICML), 2024
-
Structured Neural Networks for Density Estimation and Causal InferenceNeural Information Processing Systems (NeurIPS), 2023
-
Copula-Based Deep Survival Models for Dependent CensoringUncertainty in Artificial Intelligence (UAI), 2023
-
DuETT: Dual Event Time Transformer for Electronic Health RecordsMachine Learning for Healthcare (MLHC), 2023
-
Machine learning in computational histopathology: Challenges and OpportunitiesGenes Cells and Chromosomes, 2023
-
A Learning Based Hypothesis Test for Harmful Covariate ShiftInternational Conference on Learning Representations (ICLR), 2023
-
Anamnesic Neural Differential Equations with Orthogonal Polynomial ProjectionsInternational Conference on Learning Representations (ICLR), 2023
-
Partial Identification with Implicit Generative ModelsNeural Information Processing Systems (NeurIPS), 2022(Spotlight)
-
HiCu: Leveraging Hierarchy for Curriculum Learning in Automated ICD CodingMachine Learning for Healthcare (MLHC), 2022
-
Large Images as Long Documents: Hierarchical ViTs with Self-Supervised Pretraining in Gigapixel Image PyramidsComputer Vision and Pattern Recognition (CVPR), 2022(Oral)
-
Hierarchical Optimal Transport for Comparing Histopathology DatasetsMedical Imaging with Deep Learning (MIDL), 2022
-
Using Time-Series Privileged Information for Provably Efficient Learning of Prediction ModelsArtificial Intelligence and Statistics (AISTATS), 2022
-
Clustering Interval-Censored Time-Series for Disease PhenotypingAssociation for the Advancement of Artificial Intelligence (AAAI), 2022
-
Mitigating bias in estimating epidemic severity due to heterogeneity of epidemic on-set and data aggregationAnnals of Epidemiology (In Press), 2021
-
Neural Pharmacodynamic State Space ModelingInternational Conference on Machine Learning (ICML), 2021
-
Max-Margin learning with the Bayes factorUncertainty in Artificial Intelligence (UAI), 2018
-
Representation Learning Approaches to Detect False Arrhythmia Alarms from ECG DynamicsMachine Learning for Healthcare (MLHC), 2018
-
Variational Autoencoders for Collaborative FilteringWorld Wide Web Conference (WWW), 2018
-
On the challenges of learning with inference networks on sparse, high-dimensional dataArtificial Intelligence and Statistics (AISTATS), 2018
-
Structured Inference Networks for Nonlinear State Space ModelsAssociation for the Advancement of Artificial Intelligence (AAAI), 2017(Oral)
-
Barrier Frank-Wolfe for Marginal InferenceNeural Information Processing Systems (NeurIPS), 2015