Uri Shaham | Bar Ilan University

About Me

I am an assistant professor at the department of computer science at Bar Ilan university. I received a Ph.D. in statistics from Yale University in 2017, under the supervision of Prof. Ronald Coifman (Math), Prof. Sahand Negahban (Statistics), and Prof. Yuval Kluger (Computational Biology).

Between my graduation and 2023, I was an assistant professor adjunct at the center for outcome research and evaluation at Yale University. Before and after my Ph.D. studies, I worked for several years in the industry in various research, algorithms design, and advisory board roles. I also actively consult to several start-up companies in the fields of Machine Learning and Data Science.

Apart from my academic endeavors, until very recently I was also a commander of an infantry battalion in reserves in the Israeli Defense Force, and I am an amateur jazz pianist. When time permits, I also like to run marathons.

Research

Research Interests

My main area of interest is in developing methods for machine learning. In particular, a primary focus of mine lies in developing representation learning tools for understanding data, its latent properties, and extraction of meaningful information from it.

During the last ten years, my primary technical playground is deep learning, and my group works on a wide range of topics, such as multimodal learning, reinforcement learning, causal inference, and generalizable spectral methods. Two current major research interests of mine are multimodal representation learning from unpaired data, and causal reinforcement learning. Whenever possible, I aim to base my work on a solid mathematical and theoretical basis, and to contribute to the theoretical understanding of machine learning and deep learning.

Publications

Accepted Publications

•
Turning Sand to Gold: Recycling Data to Bridge On-Policy and Off-Policy Learning via Causal Bound
Fiskus, Tal, and Shaham, Uri.

NeurIPS 2025
•
Learning Shared Representations from Unpaired Data
Yacobi, Amitai, Ben-Ari, Nir, Talmon, Ronen, and Shaham, Uri.

NeurIPS 2025
•
Generalizable Spectral Embedding with an Application to UMAP
Ben-Ari, Nir, Yacobi, Amitai, and Shaham, Uri.

TMLR 2025
•
P-CAFE: Personalized Cost-Aware Feature Selection for Electronic Health Records
Kashani, Naama, Cohen, Mira and Shaham, Uri.

ECAI 2025
•
Generalizable and Robust Spectral Method for Multi-view Representation Learning
Yacobi, Amitai , Lindenbaum, Ofir and Shaham, Uri.

TMLR 2025
•
Supervised Domain Adaptation Based on Marginal and Conditional Distributions Alignment
Katz, Ori , Talmon, Ronen and Shaham, Uri.

TMLR 2024
•
Sparse Binarization for Fast Keyword Spotting
Svirsky, Jonathan, Shaham, Uri, and Lindenbaum, Ofir.

Interspeech 2024
•
Discovery of single independent latent variable
Shaham, Uri, Svirsky, Jonathan, Katz, Ori and Talmon, Ronen.

NeurIPS 2022
•
Deep Unsupervised Feature Selection by Discarding Nuisance and Correlated Features
Shaham, Uri, Lindenbaum, Ofir, Svirsky, Jonathan and Kluger, Yuval.

Neural Networks 2022
•
Differentiable Unsupervised Feature Selection based on a Gated Laplacian
Lindenbaum, Ofir, Shaham, Uri, Svirski, Jonathan, Peterfreund, Erez, and Kluger, Yuval.

NeurIPS 2021
•
Learning to Ask Medical Questions using Reinforcement Learning
Shaham Uri, Zahavy Tom, Caraballo Cesar, Mahajan Shiwani, Massey Daisy, and Krumholz Harlan.

Machine Learning in Healthcare 2020
•
DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network
Katzman, Jared, Shaham, Uri, Cloninger, Alexander, Bates, Jonathan, Jiang, Tingting, and Kluger, Yuval.

BMC Medical Research Methodology 2018
•
Learning by Coincidence: Siamese Networks and Common Variable Learning
Shaham, Uri, and Lederman, Roy.

Pattern Recognition 2018
•
SpectralNet: Spectral Clustering using Deep Neural Networks
Shaham, Uri, Stanton, Kelly, Li, Henry, Basri, Ronen, Nadler, Boaz, and Kluger, Yuval.

ICLR 2018
•
Understanding Adversarial Training: Increasing Local Stability of Neural Nets through Robust Optimization
Shaham, Uri, Yamada, Yutaro, and Negahban, Sahand.

Neurocomputing 2018
•
Removal of Batch Effects using Distribution-Matching Residual Networks
Shaham, Uri, Stanton, Kelly P., Zhao, Jun, Li, Huamin, Raddassi, Khadir, Montgomery, Ruth, and Kluger, Yuval.

Bioinformatics 2017
•
Gating Mass Cytometry Data by Deep Learning
Li, Huamin, Shaham, Uri, Yao, Yi, Montgomery, Ruth and Kluger, Yuval.

Bioinformatics 2017
•
Diffusion Nets
Mishne, Gal, Shaham, Uri, Cloninger, Alexander, and Cohen, Israel.

Applied and Computational Harmonic Analysis 2017
•
Provable Approximation Properties for Deep Neural Networks
Shaham, Uri, Cloninger Alexander, and Coifman Ronald R.

Applied and Computational Harmonic Analysis 2016
•
A Deep Learning Approach to Unsupervised Ensemble Learning
Shaham, Uri, Cheng, Xiuyuan, Dror, Omer, Jaffe, Ariel, Nadler, Boaz, Chang, Joseph and Kluger, Yuval.

ICML 2016

Pre-prints

•
Stochastic Neighbor Embedding Separates Well-Separated Clusters
Shaham, Uri, and Steinberger Stefan.
•
Methods for detecting co-mutated pathways in cancer samples to inform treatment selection
Jiang, Tingting, Shaham, Uri, Parisi, Fabio, Halaban, Ruth, Safonov, Anton, Kluger, Harriett, Weissman, Weismann, Chang, Joseph and Kluger. Yuval.
•
Deep Neural Network to Predict Local Failure Following Stereotactic Body Radiation Therapy: Integrating Imaging and Clinical Data to Predict Outcomes
Aneja, Sanjay, Shaham, and Krumholz, Harlan.
•
Defending against Adversarial Attacks using Basis Functions Transformations
Shaham, Uri, Garritano, Jim, Yamada, Yutaro, Weinberger, Ethan, Cloninger, Alex, Cheng, Xiuyuan, Stanton, Kelly and Kluger, Yuval.
•
Batch Effect Removal via Batch Free Encoding
Shaham, Uri.
•
Automated Characterization of Stenosis in Invasive Coronary Angiography Images with Convolutional Neural Networks
Au, Benjamin, Shaham, Uri, Dhruva, Sanket, Bouras, Georgios, Cristea, Ecaterina, Coppi, Andreas, Warner, Fred, Li, Shu-Xia, and Krumholz, Harlan.
•
Deep Ordinal Regression using Optimal Transport Loss and Unimodal Output Probabilities
Shaham, Uri and Svirsky, Jonathan.

Teaching

Seminar on Representation Learning

Fall 2025

Mathematical Methods for Data Science

Spring 2026

Machine Learning

Spring 2025

Students

Ph.D. Students

Tal Fiskus

M.Sc. Students

Past Students

A Note to Prospective Students

Are you curious about what data can tell us? Are you humble and eager to learn? Do you enjoy challenging yourself? Are you mentally strong enough to continue trying where most others give up? Do you belong to the top of your class in math? Are you a very good programmer? Are you willing to make a trip to the edge of your capabilities and then extend them further?

I am accepting a few M.Sc. and Ph.D. students. Prospective students should be available for full-time research, and will be required to work on campus at least 3 days a week. If you're interested, please email me, tell me about your interests, and attach your CV and transcript.