Scholar

Tao Sun

Google Scholar ID: fPNZpAe5WXIC

National University of Defense Technology

machine learning

Homepage↗Google Scholar↗

Citations & Impact

All-time

Citations

2,170

H-index

i10-index

Publications

Co-authors

Contact

Emailsuntao.saltfish@outlook.com GitHubOpen ↗

Publications

6 items

FedACT: Federated Adaptive Coordinate Trust Modulation for Robust Transformer Training under Data Heterogeneity

2026

Cited

Unveiling High-Probability Generalization in Decentralized SGD

2026

Cited

Revealing Modular Gradient Noise Imbalance in LLMs: Calibrating Adam via Signal-to-Noise Ratio

2026

Cited

Stability and Generalization for Decentralized Markov SGD

2026

Cited

MAny: Merge Anything for Multimodal Continual Instruction Tuning

2026

Cited

Local Gradient Regulation Stabilizes Federated Learning under Client Heterogeneity

arXiv.org · 2026

Cited

Resume

Academic Achievements

2025 ICML Conference Paper: 'Targeted Low-rank Refinement: Enhancing Sparse Language Model with Precision'
2025 CVPR Conference Paper: 'Investigating the Role of Weight Decay in Enhancing Nonconvex SGD'
2024 NeurIPS Conference Paper: 'Stability and Generalization of Asynchronous SGD: Sharper Bounds Beyond Lipschitz and Smoothness'
2024 ICML Conference Paper: 'Stability and Generalization for Stochastic Recursive Momentum-based Algorithms for (Strongly-) Convex One to K-Level Stochastic Optimizations'
2024 IJCAI Conference Paper: 'Exploring the Inefficiency of Heavy Ball as Momentum Parameter Approaches 1'
2023 ICML Conference Paper: 'Momentum Ensures Convergence of SIGNSGD under Weaker Assumptions'
2023 AAAI Conference Paper: 'Stability-Based Generalization Analysis of the Asynchronous Decentralized SGD'
2022 NeurIPS Conference Paper: 'Finite-Time Analysis of Adaptive Temporal Difference Learning with Deep Neural Networks'
2022 ICML Conference Paper: 'Adaptive Random Walk Gradient Descent for Decentralized Optimization'
2021 AAAI Conference Paper: 'Stability and Generalization of the Decentralized Stochastic Gradient Descent'
2019 IJCAI Conference Paper: 'Heavy-ball Algorithms Always Escape Saddle Points'
2019 AAAI Conference Paper: 'Non-ergodic Convergence Analysis of Heavy-ball Algorithms'
2019 NeurIPS Conference Paper: 'General Proximal Incremental Aggregated Gradient Algorithms: Better and Novel Results under General Scheme'
2018 NeurIPS Conference Paper: 'LAG: Lazily Aggregated Gradient for Communication-Efficient Distributed Learning'

Background

Associate Professor, College of Computer Science and Technology, National University of Defense Technology. Research interests include machine learning, deep learning, optimization, and distributed learning.

Miscellany

Personal interests not mentioned