Kuangyu Ding

11 papers Journal 11
YearRankTypeTitle / Venue / Authors
2025 J jnl
A New Decomposition Paradigm for Graph-structured Nonlinear Programs via Message Passing.
CoRR
Kuangyu Ding, Marie Maros, Gesualdo Scutari
2025 J jnl
Adam-family Methods with Decoupled Weight Decay in Deep Learning.
Trans. Mach. Learn. Res.
Kuangyu Ding, Nachuan Xiao, Kim-Chuan Toh
2025 J jnl
Nonconvex Stochastic Bregman Proximal Gradient Method with Application to Deep Learning.
J. Mach. Learn. Res.
Kuangyu Ding, Jingyang Li, Kim-Chuan Toh
2025 J jnl
On exploration of an interior mirror descent flow for stochastic nonconvex constrained problem.
CoRR
Kuangyu Ding, Kim-Chuan Toh
2025 J jnl
Stochastic Bregman Subgradient Methods for Nonsmooth Nonconvex Optimization Problems.
J. Optim. Theory Appl.
Kuangyu Ding, Kim-Chuan Toh
2024 J jnl
Developing Lagrangian-based Methods for Nonsmooth Nonconvex Optimization.
CoRR
Nachuan Xiao, Kuangyu Ding, Xiaoyin Hu, Kim-Chuan Toh
2024 J jnl
Memory-Efficient 4-bit Preconditioned Stochastic Optimization.
CoRR
Jingyang Li, Kuangyu Ding, Kim-Chuan Toh, Pan Zhou
2024 J jnl
Optimization Hyper-parameter Laws for Large Language Models.
CoRR
Xingyu Xie, Kuangyu Ding, Shuicheng Yan, Kim-Chuan Toh, Tianwen Wei
2023 J jnl
Adam-family Methods with Decoupled Weight Decay in Deep Learning.
CoRR
Kuangyu Ding, Nachuan Xiao, Kim-Chuan Toh
2023 J jnl
Nonconvex Stochastic Bregman Proximal Gradient Method with Application to Deep Learning.
CoRR
Kuangyu Ding, Jingyang Li, Kim-Chuan Toh
2023 J jnl
On proximal augmented Lagrangian based decomposition methods for dual block-angular convex composite programming problems.
Comput. Optim. Appl.
Kuangyu Ding, Xin Yee Lam, Kim-Chuan Toh