Research | Xuyang Wu

Historical Information-based Optimization

This topic studies how historical iterates, gradients, and other algorithm information can be reused to improve convergence of optimization algorithms.

Anderson Accelerated Asynchronous Method for Distributed Optimization TMLR

C. Li and X. Wu. Transactions on Machine Learning Research, 2026.

[paper]
Anderson Acceleration for Distributed Constrained Optimization over Time-varying Networks Preprint

H. Liu and X. Wu. Preprint, 2026.

[paper]
Historical Information Accelerates Decentralized Optimization: A Proximal Bundle Method Preprint

Z. Zhu, Y.-P. Tian, and X. Wu. Preprint, 2025.

[paper]
Primal-Dual Bundle Methods for Linear Equality-Constrained Problems Preprint

Z. Zheng, T. Liu, and X. Wu. Preprint, 2025.

[paper]
An Accelerated Proximal Bundle Method with Momentum CDC

Z. Zheng, J. Yin, S. Yang, and X. Wu. IEEE Conference on Decision and Control, 2026.

[paper]
Bundle EXTRA for Decentralized Optimization CDC

H. Liu, Z. Zheng, C. Li, W. Xu, and X. Wu. IEEE Conference on Decision and Control, 2026.

[paper]
An Asynchronous Bundle Method for Distributed Learning Problems ICLR

D. Cederberg, X. Wu, S. P. Boyd, and M. Johansson. International Conference on Learning Representations, 2025.

[paper]

Asynchronous Optimization

We develop delay-robust methods for optimization and learning under heterogeneous computation and communication.

Anderson Accelerated Asynchronous Method for Distributed Optimization TMLR

C. Li and X. Wu. Transactions on Machine Learning Research, 2026.

[paper]
An Asynchronous Bundle Method for Distributed Learning Problems ICLR

D. Cederberg, X. Wu, S. P. Boyd, and M. Johansson. International Conference on Learning Representations, 2025.

[paper]
Asynchronous Distributed Optimization with Delay-free Parameters TAC

X. Wu, C. Liu, S. Magnusson, and M. Johansson. IEEE Transactions on Automatic Control, 2025.

[paper]
Delay-agnostic Asynchronous Coordinate Update Algorithm ICML

X. Wu, C. Liu, S. Magnusson, and M. Johansson. International Conference on Machine Learning, 2023.

[paper]
Delay-adaptive Step-sizes for Asynchronous Learning ICML

X. Wu, H. R. Feyzmahdavian, S. Magnusson, and M. Johansson. International Conference on Machine Learning, 2022.

[paper]
Optimal Convergence Rates of Totally Asynchronous Optimization CDC

X. Wu, H. R. Feyzmahdavian, S. Magnusson, and M. Johansson. IEEE Conference on Decision and Control, 2022.

[paper]

Stochastic Optimization

This direction concerns optimization methods for learning and large-scale problems involving stochastic updates.

Historical Information Accelerates Decentralized Optimization: A Proximal Bundle Method Preprint

Z. Zhu, Y.-P. Tian, and X. Wu. Preprint, 2025.

[paper]
From Promise to Practice: Realizing High-performance Decentralized Training ICLR

Z. Wang, J. Zhang, X. Wu, and M. Johansson. International Conference on Learning Representations, 2025.

[paper]
An Asynchronous Bundle Method for Distributed Learning Problems ICLR

D. Cederberg, X. Wu, S. P. Boyd, and M. Johansson. International Conference on Learning Representations, 2025.

[paper]

Decentralized Optimization

We develop communication-efficient algorithms to solve optimization problems over networks.

From Promise to Practice: Realizing High-performance Decentralized Training ICLR

Z. Wang, J. Zhang, X. Wu, and M. Johansson. International Conference on Learning Representations, 2025.

[paper]
Anderson Acceleration for Distributed Constrained Optimization over Time-varying Networks Preprint

H. Liu and X. Wu. Preprint, 2026.

[paper]
Historical Information Accelerates Decentralized Optimization: A Proximal Bundle Method Preprint

Z. Zhu, Y.-P. Tian, and X. Wu. Preprint, 2025.

[paper]
Bundle EXTRA for Decentralized Optimization CDC

H. Liu, Z. Zheng, C. Li, W. Xu, and X. Wu. IEEE Conference on Decision and Control, 2026.

[paper]
A Unified Dual Consensus Approach to Distributed Optimization with Globally-Coupled Constraints TCNS

Z. Liu, X. Wu, D. Wang, and J. Lu. IEEE Transactions on Control of Networked Systems, 2025.

[paper]
Globally-Constrained Decentralized Optimization with Variable Coupling TAC

D. Wang, X. Wu, Z. Ou, and J. Lu. IEEE Transactions on Automatic Control, 2025.

[paper]
Achieving Violation-free Distributed Optimization under Coupling Constraints Automatica

C. Liu, X. Tan, X. Wu, D. V. Dimarogonas, and K. H. Johansson. Automatica, 2025.

[paper]
A Regularized Fenchel Dual Gradient Method for Nonsmooth Optimization over Time-varying Networks OMS

X. Wu, K. Sou, and J. Lu. Optimization Methods and Software, 2023.

[paper]
Distributed Safe Resource Allocation using Barrier Functions Automatica

X. Wu, S. Magnusson, and M. Johansson. Automatica, 2023.

[paper]
Distributed Optimization with Coupling Constraints TAC

X. Wu, H. Wang, and J. Lu. IEEE Transactions on Automatic Control, 2023.

[paper]
A Unifying Approximate Method of Multipliers for Distributed Composite Optimization TAC

X. Wu and J. Lu. IEEE Transactions on Automatic Control, 2023.

[paper]
A Second-Order Proximal Algorithm for Consensus Optimization TAC

X. Wu, Z. Qu, and J. Lu. IEEE Transactions on Automatic Control, 2021.

[paper]
Distributed Optimization over Time-varying Networks with Minimal Connectivity L-CSS

X. Wu and J. Lu. IEEE Control Systems Letters, 2020.

[paper]
Fenchel Dual Gradient Methods for Distributed Convex Optimization over Time-varying Networks TAC

X. Wu and J. Lu. IEEE Transactions on Automatic Control, 2019.

[paper]
Rate Analysis of Dual Averaging for Nonconvex Distributed Optimization IFAC

C. Liu, X. Wu, X. Yi, Y. Shi, and K. H. Johansson. IFAC World Congress, 2023.

[paper]

Safety-critical Optimization

We propose algorithms that produce all-feasible iterates to solve optimization problems in safety-critical scenarios.

Achieving Violation-free Distributed Optimization under Coupling Constraints Automatica

C. Liu, X. Tan, X. Wu, D. V. Dimarogonas, and K. H. Johansson. Automatica, 2025.

[paper]
Distributed Safe Resource Allocation using Barrier Functions Automatica

X. Wu, S. Magnusson, and M. Johansson. Automatica, 2023.

[paper]
A New Family of Feasible Methods for Distributed Resource Allocation CDC

X. Wu, S. Magnusson, and M. Johansson. IEEE Conference on Decision and Control (invited), 2021.

[paper]

Optimization over Time-varying Networks

We solve optimization problems over networks that change with iterations.

Anderson Acceleration for Distributed Constrained Optimization over Time-varying Networks Preprint

H. Liu and X. Wu. Preprint, 2026.

[paper]
A Regularized Fenchel Dual Gradient Method for Nonsmooth Optimization over Time-varying Networks OMS

X. Wu, K. Sou, and J. Lu. Optimization Methods and Software, 2023.

[paper]
Distributed Optimization over Time-varying Networks with Minimal Connectivity L-CSS

X. Wu and J. Lu. IEEE Control Systems Letters, 2020.

[paper]
Fenchel Dual Gradient Methods for Distributed Convex Optimization over Time-varying Networks TAC

X. Wu and J. Lu. IEEE Transactions on Automatic Control, 2019.

[paper]