Ruoyu Sun

Affiliations

Chinese University of Hong Kong, Shenzhen

papers

574

total citations

papers (16)

ReMax: A Simple, Effective, and Efficient Reinforcement Learning Method for Aligning Large Language Models

ICML 2024arXiv

147

citations

A Single-Loop Smoothed Gradient Descent-Ascent Algorithm for Nonconvex-Concave Min-Max Problems

NEURIPS 2020arXiv

118

citations

Adam Can Converge Without Any Modification On Update Rules

NEURIPS 2022arXiv

citations

Global Convergence of MAML and Theory-Inspired Neural Architecture Search for Few-Shot Learning

CVPR 2022arXiv

citations

DigGAN: Discriminator gradIent Gap Regularization for GAN Training with Limited Data

NEURIPS 2022arXiv

citations

PDHG-Unrolled Learning-to-Optimize Method for Large-Scale Linear Programming

ICML 2024arXiv

citations

Balanced Training for Sparse GANs

NEURIPS 2023arXiv

citations

When Expressivity Meets Trainability: Fewer than $n$ Neurons Can Work

NEURIPS 2021arXiv

citations

PAC-Bayesian Spectrally-Normalized Bounds for Adversarially Robust Generalization

NEURIPS 2023arXiv

citations

How Graph Neural Networks Learn: Lessons from Training Dynamics

ICML 2024arXiv

citations

When GNNs meet symmetry in ILPs: an orbit-based feature augmentation approach

ICLR 2025arXiv

citations

Towards Explaining the Power of Constant-depth Graph Neural Networks for Structured Linear Programming

ICLR 2025

citations

Faster Directional Convergence of Linear Neural Networks under Spherically Symmetric Data

NEURIPS 2021

citations

Ruoyu Sun

Affiliations

papers (16)

ReMax: A Simple, Effective, and Efficient Reinforcement Learning Method for Aligning Large Language Models

A Single-Loop Smoothed Gradient Descent-Ascent Algorithm for Nonconvex-Concave Min-Max Problems

Adam Can Converge Without Any Modification On Update Rules

Global Convergence of MAML and Theory-Inspired Neural Architecture Search for Few-Shot Learning

Stability Analysis and Generalization Bounds of Adversarial Training

Towards a Better Global Loss Landscape of GANs

Does Momentum Change the Implicit Regularization on Separable Data?

DigGAN: Discriminator gradIent Gap Regularization for GAN Training with Limited Data

PDHG-Unrolled Learning-to-Optimize Method for Large-Scale Linear Programming

Balanced Training for Sparse GANs

When Expressivity Meets Trainability: Fewer than $n$ Neurons Can Work

PAC-Bayesian Spectrally-Normalized Bounds for Adversarially Robust Generalization

How Graph Neural Networks Learn: Lessons from Training Dynamics

When GNNs meet symmetry in ILPs: an orbit-based feature augmentation approach

Towards Explaining the Power of Constant-depth Graph Neural Networks for Structured Linear Programming

Faster Directional Convergence of Linear Neural Networks under Spherically Symmetric Data

papers (16)

ReMax: A Simple, Effective, and Efficient Reinforcement Learning Method for Aligning Large Language Models

A Single-Loop Smoothed Gradient Descent-Ascent Algorithm for Nonconvex-Concave Min-Max Problems

Adam Can Converge Without Any Modification On Update Rules

Global Convergence of MAML and Theory-Inspired Neural Architecture Search for Few-Shot Learning

Stability Analysis and Generalization Bounds of Adversarial Training

Towards a Better Global Loss Landscape of GANs

Does Momentum Change the Implicit Regularization on Separable Data?

DigGAN: Discriminator gradIent Gap Regularization for GAN Training with Limited Data

PDHG-Unrolled Learning-to-Optimize Method for Large-Scale Linear Programming

Balanced Training for Sparse GANs

When Expressivity Meets Trainability: Fewer than $n$ Neurons Can Work

PAC-Bayesian Spectrally-Normalized Bounds for Adversarially Robust Generalization

How Graph Neural Networks Learn: Lessons from Training Dynamics

When GNNs meet symmetry in ILPs: an orbit-based feature augmentation approach

Towards Explaining the Power of Constant-depth Graph Neural Networks for Structured Linear Programming

Faster Directional Convergence of Linear Neural Networks under Spherically Symmetric Data