Mikhail Belkin

Professor, University of California, San Diego

Joined

February 2019

Names

Mikhail Belkin (Preferred)

Misha Belkin

Emails

****@gmail.com (Confirmed)

****@cse.ohio-state.edu (Confirmed)

****@ucsd.edu (Confirmed)

Personal Links

Homepage

Google Scholar

Career & Education History

Professor

University of California, San Diego (ucsd.edu)

Present

Advisors, Relations & Conflicts

No relations added

Expertise

No areas of expertise listed

Publications

Neural Tangent Kernel at Initialization: Linear Width Suffices
Arindam Banerjee, Pedro Cisneros-Velarde, Libin Zhu, Misha Belkin
- Published: 08 May 2023, Last Modified: 26 Jun 2023
- UAI 2023
- Readers: Everyone
Loss landscapes and optimization in over-parameterized non-linear systems and neural networks
Chaoyue Liu, Libin Zhu, Misha Belkin
- 25 Dec 2022
- OpenReview Archive Direct Upload
- Readers: Everyone
Restricted Strong Convexity of Deep Learning Models with Smooth Activations
Arindam Banerjee, Pedro Cisneros-Velarde, Libin Zhu, Misha Belkin
- Published: 01 Feb 2023, Last Modified: 19 Apr 2023
- ICLR 2023 poster
- Readers: Everyone
Quadratic models for understanding neural network dynamics
Libin Zhu, Chaoyue Liu, Adityanarayanan Radhakrishnan, Misha Belkin
- Published: 01 Feb 2023, Last Modified: 14 Jan 2026
- Submitted to ICLR 2023
- Readers: Everyone
Learning large-scale Kernel Networks
Amirhesam Abedsoltan, Parthe Pandit, Misha Belkin
- Published: 01 Feb 2023, Last Modified: 13 Feb 2023
- Submitted to ICLR 2023
- Readers: Everyone
Benign, Tempered, or Catastrophic: Toward a Refined Taxonomy of Overfitting
Neil Rohit Mallinar, James B Simon, Amirhesam Abedsoltan, Parthe Pandit, Misha Belkin, Preetum Nakkiran
- Published: 31 Oct 2022, Last Modified: 11 Jan 2023
- NeurIPS 2022 Accept
- Readers: Everyone
Benign Overfitting in Two-layer Convolutional Neural Networks
Yuan Cao, Zixiang Chen, Misha Belkin, Quanquan Gu
- Published: 31 Oct 2022, Last Modified: 11 Jan 2023
- NeurIPS 2022 Accept
- Readers: Everyone
Transition to Linearity of General Neural Networks with Directed Acyclic Graph Architecture
Libin Zhu, Chaoyue Liu, Misha Belkin
- Published: 31 Oct 2022, Last Modified: 11 Oct 2022
- NeurIPS 2022 Accept
- Readers: Everyone
Transition to Linearity of Wide Neural Networks is an Emerging Property of Assembling Weak Models
Chaoyue Liu, Libin Zhu, Misha Belkin
- Published: 28 Jan 2022, Last Modified: 13 Feb 2023
- ICLR 2022 Spotlight
- Readers: Everyone
Risk Bounds for Over-parameterized Maximum Margin Classification on Sub-Gaussian Mixtures
Yuan Cao, Quanquan Gu, Misha Belkin
- Published: 09 Nov 2021, Last Modified: 05 May 2023
- NeurIPS 2021 Poster
- Readers: Everyone

View all 12 publications