Eric Zhao

I'm a PhD candidate in computer science at UC Berkeley, where I'm advised by Nika Haghtalab and Michael I. Jordan. I'm a Google PhD Fellow, a NSF Graduate Research Fellow, and affiliated with the Berkeley AI Research Lab (BAIR).

My research studies the foundations of multi-domain machine learning. I currently work on enabling large language models to coherently reason in new domains, especially domains of their own creation—for example, being able to invent and then prove results within a novel mathematical language. On the theoretical side of my research, I work on developing mathematical frameworks for provably obtaining multi-domain intelligence, which has also led to new results in statistical learning theory and calibrated forecasting.

I received my B.S. from Caltech in 2020 and previously interned with Google Research, Nvidia Research, Salesforce Research, and Uber.

Selected Works

(α-β) denotes when authors are ordered alphabetically.

Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification Eric Zhao, Pranjal Awasthi, Sreenivas Gollapudi
PDF Preprint, Jan 2025
A Unifying Perspective on Multi-Calibration: Game Dynamics for Multi-Objective Learning (α-β) Nika Haghtalab, Michael I. Jordan, Eric Zhao
PDF NeurIPS 2023
On-Demand Sampling: Learning Optimally from Multiple Distributions (α-β) Nika Haghtalab, Michael I. Jordan, Eric Zhao
PDF NeurIPS 2022

Selected Awards

Google PhD Fellowship (2024)
NSF Graduate Research Fellowship (2023)
NeurIPS Outstanding Paper Award (2022)

Contact

eric.zh@berkeley.edu