Zhuoli Ouyang

oyzl2004@gmail.com

ouyangzl2023@mail.sustech.edu.cn

Zhuoli.Ouyang@dartmouth.edu

I am a third-year undergraduate at SUSTech, majoring in Electrical and Electronic Engineering. I began in the Fields Elite Honor Class at the Department of Mathematics, where early immersion in rigorous proof and foundational theory fundamentally shaped my research philosophy.

I am dedicated to revitalizing centuries of profound mathematical brilliance within the AI era. Rather than chasing empirical efficiency alone, my mission is to bridge the legacy of classical theory with the frontiers of modern intelligence — grounding AI in rigorous, enduring paradigms and opening new theoretical insights.

I believe the most powerful AI systems will be those that stand on the shoulders of centuries of profound theoretical legacy. My approach centers on the rigorous rethinking of mathematical structures — prioritizing clarity and depth over sheer scale — and translating abstract mathematical elegance into actionable frameworks that offer both transparency and enduring performance.

I am currently seeking PhD positions (2027 Fall). Feel free to reach out!

Awards & Honors

ALT 2026 Best Student Paper Award
ISIT 2026 Strong Accept
DATE 2026 Oral Paper
Fields Elite Honor Class (Top 1%), SUSTech
Rank 4/46 in EEE, Top 7% overall (GPA 3.94/4.00 last year)

news

May 13, 2026	Our paper RMNP: Row-Momentum Normalized Preconditioning for Scalable Matrix-Based Optimization is accepted at ICML 2026! (co-first author) [arXiv]
May 12, 2026	Our paper Understanding Generalization from Embedding Dimension and Distributional Convergence is accepted at ICML 2026! (co-first author) [arXiv]
Feb 01, 2026	Our paper Suspicious Alignment of SGD: A Fine-Grained Step Size Condition Analysis received the Best Student Paper Award at ALT 2026! [arXiv]
Jan 15, 2026	Our paper Depth, Not Data: An Analysis of Hessian Spectral Bifurcation is accepted at ISIT 2026 (Strong Accept)! (co-first author) [arXiv]
Jan 10, 2026	Our paper Doloris: Dual Conditional Diffusion Implicit Bridges with Sparsity Masking Strategy for Unpaired Single-Cell Perturbation Estimation is accepted at ICLR 2026! [OpenReview]
Nov 15, 2025	Our paper ACEMARL: Adaptive Clustering Enhanced Multi-Agent Reinforcement Learning for Analog Circuit Sizing is accepted at DATE 2026 as an Oral presentation! [paper]
Sep 20, 2025	Our paper FD-MAGRPO: Functionality-Driven Multi-Agent Group Relative Policy Optimization for Analog-LDO Sizing is accepted at AAAI 2026! [paper]

selected publications

ICML

RMNP: Row-Momentum Normalized Preconditioning for Scalable Matrix-Based Optimization

Shenyang Deng^*, Zhuoli Ouyang^*, Tianyu Pang, Zihang Liu, Ruochen Jin, and 2 more authors

In International Conference on Machine Learning (ICML), 2026

arXiv Bib

@inproceedings{deng2026rmnp,
  title = {RMNP: Row-Momentum Normalized Preconditioning for Scalable Matrix-Based Optimization},
  author = {Deng, Shenyang and Ouyang, Zhuoli and Pang, Tianyu and Liu, Zihang and Jin, Ruochen and Yu, Shuhua and Yang, Yaoqing},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
}

ICML

Understanding Generalization from Embedding Dimension and Distributional Convergence

Junjie Yu^*, Zhuoli Ouyang^*, Haotian Deng^*, Chen Wei, Wenxiao Ma, and 3 more authors

In International Conference on Machine Learning (ICML), 2026

arXiv Bib

@inproceedings{yu2026generalization,
  title = {Understanding Generalization from Embedding Dimension and Distributional Convergence},
  author = {Yu, Junjie and Ouyang, Zhuoli and Deng, Haotian and Wei, Chen and Ma, Wenxiao and Zhang, Jianyu and Deng, Zihan and Liu, Quanying},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
}

ISIT

Depth, Not Data: An Analysis of Hessian Spectral Bifurcation

Shenyang Deng^*, Boyao Liao^*, Zhuoli Ouyang^*, Tianyu Pang, and Yaoqing Yang

In IEEE International Symposium on Information Theory (ISIT), 2026

Strong Accept arXiv Bib

Strong Accept

@inproceedings{deng2026depth,
  title = {Depth, Not Data: An Analysis of Hessian Spectral Bifurcation},
  author = {Deng, Shenyang and Liao, Boyao and Ouyang, Zhuoli and Pang, Tianyu and Yang, Yaoqing},
  booktitle = {IEEE International Symposium on Information Theory (ISIT)},
  year = {2026},
}

ALT

Suspicious Alignment of SGD: A Fine-Grained Step Size Condition Analysis

Shenyang Deng^*, Boyao Liao^*, Zhuoli Ouyang, Tianyu Pang, Minhak Song, and 1 more author

In Algorithmic Learning Theory (ALT), 2026

Best Student Paper arXiv Bib

Best Student Paper Award

@inproceedings{deng2026suspicious,
  title = {Suspicious Alignment of SGD: A Fine-Grained Step Size Condition Analysis},
  author = {Deng, Shenyang and Liao, Boyao and Ouyang, Zhuoli and Pang, Tianyu and Song, Minhak and Yang, Yaoqing},
  booktitle = {Algorithmic Learning Theory (ALT)},
  year = {2026},
}

ICLR

Doloris: Dual Conditional Diffusion Implicit Bridges with Sparsity Masking Strategy for Unpaired Single-Cell Perturbation Estimation

Changxi Chi, Jun Xia, Yufei Huang, and Zhuoli Ouyang

In International Conference on Learning Representations (ICLR), 2026

Bib HTML

@inproceedings{chi2026doloris,
  title = {Doloris: Dual Conditional Diffusion Implicit Bridges with Sparsity Masking Strategy for Unpaired Single-Cell Perturbation Estimation},
  author = {Chi, Changxi and Xia, Jun and Huang, Yufei and Ouyang, Zhuoli},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2026},
}

DATE

ACEMARL: Adaptive Clustering Enhanced Multi-Agent Reinforcement Learning for Analog Circuit Sizing

Han Wu^*, Haoning Jiang^*, Zhuoli Ouyang, Ziheng Wang, Qi Shen, and 3 more authors

In Design, Automation & Test in Europe Conference (DATE), 2026

Oral Bib HTML

Oral

@inproceedings{wu2026acemarl,
  title = {ACEMARL: Adaptive Clustering Enhanced Multi-Agent Reinforcement Learning for Analog Circuit Sizing},
  author = {Wu, Han and Jiang, Haoning and Ouyang, Zhuoli and Wang, Ziheng and Shen, Qi and Yuan, Bo and Lu, Yan and Jiang, Junmin},
  booktitle = {Design, Automation \& Test in Europe Conference (DATE)},
  year = {2026},
}

AAAI

FD-MAGRPO: Functionality-Driven Multi-Agent Group Relative Policy Optimization for Analog-LDO Sizing

Haoning Jiang^*, Han Wu^*, Zhuoli Ouyang, Ziheng Wang, Tinghuan Chen, and 1 more author

In AAAI Conference on Artificial Intelligence, 2026

Bib HTML

@inproceedings{jiang2026fdmagrpo,
  title = {FD-MAGRPO: Functionality-Driven Multi-Agent Group Relative Policy Optimization for Analog-LDO Sizing},
  author = {Jiang, Haoning and Wu, Han and Ouyang, Zhuoli and Wang, Ziheng and Chen, Tinghuan and Jiang, Junmin},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year = {2026},
}