Colin Wei

Joined

September 2018

Names

Colin Wei (Preferred)

Emails

****@stanford.edu (Confirmed)

Personal Links

Homepage

DBLP

Career & Education History

PhD student

Computer Science Department, Stanford University (cs.stanford.edu)

2017 – 2022

Advisors, Relations & Conflicts

PhD Advisor

Tengyu Ma

2017 – 2022

Expertise

theory

Present

non-convex optimization

Present

generalization theory

Present

Publications

Max-Margin Works while Large Margin Fails: Generalization without Uniform Convergence
Margalit Glasgow, Colin Wei, Mary Wootters, Tengyu Ma
- Published: 01 Feb 2023, Last Modified: 01 Mar 2023
- ICLR 2023 poster
- Readers: Everyone
Statistically Meaningful Approximation: a Case Study on Approximating Turing Machines with Transformers
Colin Wei, Yining Chen, Tengyu Ma
- Published: 31 Oct 2022, Last Modified: 21 Oct 2022
- NeurIPS 2022 Accept
- Readers: Everyone
Beyond Separability: Analyzing the Linear Transferability of Contrastive Representations to Related Subpopulations
Jeff Z. HaoChen, Colin Wei, Ananya Kumar, Tengyu Ma
- Published: 31 Oct 2022, Last Modified: 11 Jan 2023
- NeurIPS 2022 Accept
- Readers: Everyone
Statistically Meaningful Approximation: a Theoretical Analysis for Approximating Turing Machines with Transformers
Colin Wei, Yining Chen, Tengyu Ma
- Published: 28 Jan 2022, Last Modified: 13 Feb 2023
- ICLR 2022 Submitted
- Readers: Everyone
Certified Robustness for Deep Equilibrium Models via Interval Bound Propagation
Colin Wei, J Zico Kolter
- Published: 28 Jan 2022, Last Modified: 12 Oct 2025
- ICLR 2022 Poster
- Readers: Everyone
Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Loss
Jeff Z. HaoChen, Colin Wei, Adrien Gaidon, Tengyu Ma
- Published: 09 Nov 2021, Last Modified: 05 May 2023
- NeurIPS 2021 Oral
- Readers: Everyone
Why Do Pretrained Language Models Help in Downstream Tasks? An Analysis of Head and Prompt Tuning
Colin Wei, Sang Michael Xie, Tengyu Ma
- Published: 09 Nov 2021, Last Modified: 26 May 2025
- NeurIPS 2021 Spotlight
- Readers: Everyone
Meta-learning Transferable Representations with a Single Target Domain
Hong Liu, Jeff Z. HaoChen, Colin Wei, Tengyu Ma
- 28 Sept 2020 (modified: 12 Oct 2025)
- ICLR 2021 Conference Blind Submission
- Readers: Everyone
Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data
Colin Wei, Kendrick Shen, Yining Chen, Tengyu Ma
- Published: 12 Jan 2021, Last Modified: 12 Oct 2025
- ICLR 2021 Oral
- Readers: Everyone
Shape Matters: Understanding the Implicit Bias of the Noise Covariance
Jeff Z. HaoChen, Colin Wei, Jason D. Lee, Tengyu Ma
- 28 Sept 2020 (modified: 12 Oct 2025)
- ICLR 2021 Conference Blind Submission
- Readers: Everyone

View all 18 publications

Colin Wei

Names

Emails

Personal Links

Career & Education History

Advisors, Relations & Conflicts

Expertise

Publications

Co-Authors