I’m a fourth year PhD student in Harvard’s Theory of Computation and Machine Learning Foundations groups. I’m advised by Leslie Valiant and Sham Kakade, and supported by an NSF Graduate Research Fellowship. Previously, I undergraduated in math at Princeton.
In my research I study how deep learning works and how learning interacts with strategic behavior. I’m also interested in theoretical computer science broadly.
bedelman@g.harvard.edu | Google Scholar | he/him
News
May 2021 Our paper on the Colonel Blotto Game was accepted to Games and Economics Behavior.
April 2021 I am interning with the ML group at Microsoft Research NYC this summer, mentored by Cyril Zhang and Sham Kakade.
March 2021 Chara Podimata, Yo Shavit, and I organized a tutorial at FAccT 2021 called How to Achieve Both Transparency and Accuracy in Predictive Decision Making: An Introduction to Strategic Prediction.
Research
Inductive Biases and Variable Creation in Self-Attention Mechanisms
with Surbhi Goel, Sham Kakade, and Cyril Zhang
Manuscript
The Multiplayer Colonel Blotto Game
with Enric Boix-Adserà and Siddhartha Jayanti
Games and Economic Behavior (full version), EC 2020 (extended abstract)
Causal Strategic Linear Regression
with Yonadav Shavit and Brian Axelrod
ICML 2020
SGD on Neural Networks Learns Functions of Increasing Complexity
with Preetum Nakkiran, Gal Kaplun, Dimitris Kalimeris, Tristan Yang, Fred Zhang, and Boaz Barak
NeurIPS 2019 (Spotlight)
Matrix Rigidity and the Croot-Lev-Pach Lemma
with Zeev Dvir
Theory of Computing, 2019
Teaching
Spring 2021 Teaching fellow for CS 229br: Biology and Complexity (taught by Leslie Valiant)
Spring 2020 Teaching fellow for CS 228: Computational Learning Theory (taught by Leslie Valiant)