Idan Mehalel

I am a postdoctoral researcher at the NSF/Simons Collaboration on the Theoretical Foundations of Deep Learning, where I am advised by Amit Daniely (HUJI) and Elchanan Mossel (MIT). Prior to that, I was a PhD student at the Technion, advised by Yuval Filmus and Shay Moran. I am working primarily on the Foundations of Machine Learning. You can check out my c.v. here.

Contact information

Email: idanmehalel@gmail.com

Office: A435, School of Computer Science and Engineering, HUJI.

Click the bullet for additional relevant links.

Amit Daniely and Idan Mehalel. Most Convolutional Networks Suffer from Small Adversarial Perturbations. 2026.

Click the bullet for additional relevant links.

Amit Daniely, Idan Mehalel and Elchanan Mossel. Online Learning of Neural Networks. NeurIPS 2025.

Zachary Chase and Idan Mehalel. Deterministic Apple Tasting. COLT 2025.

Yuval Filmus, Steve Hanneke, Idan Mehalel and Shay Moran. Bandit-Feedback Online Multiclass Classification: Variants and Tradeoffs. NeurIPS 2024.

Idan Mehalel, Ananth Raman, Vinod Raman, Unique Subedi and Ambuj Tewari. Multiclass Online Learnability Under Bandit Feedback. ALT 2024.

Steve Hanneke, Amin Karbasi, Mohammad Mahmoody, Idan Mehalel and Shay Moran. On Optimal Learning Under Targeted Data Poisoning. NeurIPS 2022.
Oral Presentation

Yuval FIlmus, Idan Mehalel and Shay Moran. A Resiliant Distributed Boosting Algorithm. ICML 2022.

Click the bullet for additional relevant links.

Yuval Filmus, Steve Hanneke, Idan Mehalel and Shay Moran. Optimal Prediction using Expeft Advice and Randomized Littlestone
DImension. SIAM Journal on Computing (SICOMP), 2025.

Yuval Filmus and Idan Mehalel. Optimal sets of questions for Twenty Questions. SIAM Journal on Discrete Mathematics (SIDMA), 2024.

Technion Computer Science department research excellence scholarship, Spring 2023.