Sotabase

Career

· Researcher in Multi-Agent Reinforcement Learning, MIT2026–
· Researcher, MIT Vision Group2025–
· Author of academic publications, MIT2023–
· PhD in Computer Science, Princeton University2022–
· Postdoctoral Fellow in Computer Science, Stanford University2022–
· Assistant Professor, Toyota Technological Institute at Chicago2022–
· Undergraduate Student, Yao Class, Tsinghua University2022–
· Researcher (visiting faculty), Google Research

Publications (0)

Fine-Grained Analysis of Optimization and Generalization for Overparameterized Two-Layer Neural Networks
International Conference on Machine Learning · 2019
1,030
cited
On Exact Computation with an Infinitely Wide Neural Net
Neural Information Processing Systems · 2019
997
cited
First Sagittarius A* Event Horizon Telescope Results. VI. Testing the Black Hole Metric
Astrophysical Journal Letters · 2022
545
cited
Towards Understanding the Role of Over-Parametrization in Generalization of Neural Networks
arXiv.org · 2018
396
cited
Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training
International Conference on Learning Representations · 2023
236
cited
The role of over-parametrization in generalization of neural networks
International Conference on Learning Representations · 2019
206
cited
Harnessing the Power of Infinitely Wide Deep Nets on Small-data Tasks
International Conference on Learning Representations · 2019
166
cited
Constraints on black-hole charges with the 2017 EHT observations of M87*
Physical Review D · 2021
139
cited
Theoretical Analysis of Auto Rate-Tuning by Batch Normalization
International Conference on Learning Representations · 2018
139
cited
Enhanced Convolutional Neural Tangent Kernels
arXiv.org · 2019
136
cited
Understanding Gradient Descent on Edge of Stability in Deep Learning
arXiv.org · 2022
125
cited
What Happens after SGD Reaches Zero Loss? -A Mathematical Framework
International Conference on Learning Representations · 2021
115
cited
Explaining Landscape Connectivity of Low-cost Solutions for Multilayer Nets
Neural Information Processing Systems · 2019
102
cited
Efficient collection of sensor data in remote fields using mobile collectors
Proceedings. 13th International Conference on Computer Communications and Networks (IEEE Cat. No.04EX969) · 2004
91
cited
Understanding the Generalization Benefit of Normalization Layers: Sharpness Reduction
Neural Information Processing Systems · 2022
89
cited
Event Horizon Telescope observations of the jet launching and collimation in Centaurus A
Nature Astronomy · 2021
88
cited
Gradient Descent on Two-layer Nets: Margin Maximization and Simplicity Bias
Neural Information Processing Systems · 2021
84
cited
Polarimetric Properties of Event Horizon Telescope Targets from ALMA
Astrophysical Journal Letters · 2021
71
cited
Dynamic Resource Discovery Based on Preference and Movement Pattern Similarity for Large-Scale Social Internet of Things
IEEE Internet of Things Journal · 2016
70
cited
The persistent shadow of the supermassive black hole of M 87
Astronomy & Astrophysics · 2024
61
cited
Sotabase