Xin Liu

Xin Liu is a tenure-track assistant professor at the School of Information Science and Technology at the ShanghaiTech University. He was a postdoctoral research fellow at the Electrical Engineering and Computer Science Department of the University of Michigan, Ann Arbor, working with Prof. Lei Ying. He received the Ph.D. degree in Electrical Engineering at Arizona State University, advised by Prof. Lei Ying in 2019. He received Master's degree in Signal and Information Processing at University of Chinese Academy of Sciences in 2014, and Bachelor's degree in Electrical Engineering at Hunan University in 2011.

Email  /  Google Scholar  /  CV

Immediate Openings: I am looking for self-motivated (undergraduate, graduate, visiting) students and postdocs who are interested in networking, learning, and optimization. Please email me if you are interested.

Recent News
- 05/2024: A paper on Safe Bandit Learning via Lyapunov Opt Based E2D got accepted at COLT 2024!
- 01/2024: A paper on Bandit Learning with Abandonment got accepted at Journal of Machine Learning Research!
- 12/2023: A paper on Bandit Learning for Online Task Scheduling got accepted at AAMAS 2024 (Oral)!
- 12/2022: Invited to serve on the TPC of Mobihoc 2024 and ICDCS 2024.
- 12/2023: A paper on Aggressive Exploration for Safe RL got accepted at AAAI 2024 (Oral)!
- 11/2023: A paper on Zero-waiting Load Balancing got accepted at the prestigious Stochastic Systems Journal!
- 09/2023: A paper on Augmented Samples for Mixed MDP got accepted at NeurIPS 2023 (Spotlight)!
- 07/2023: A paper on Optimal Microservice Configuration got accepted at Performance 2023!

Research Interests

My research lies broadly in stochastic modeling, analysis, and optimization, online learning and decision-making, and reinforcement learning with applications in large-scale server systems, communication and ride-sharing networks, etc.

Publications

Conference Articles

Stochastic Constrained Contextual Bandits via Lyapunov Optimization Based Estimation to Decision Framework.
COLT 2024
Hengquan Guo and Xin Liu

Learning to Schedule Online Tasks with Bandit Feedback.
AAMAS 2024
Yongxin Xu, Shangshang Wang, Hengquan Guo, Xin Liu, and Ziyu Shao

Latency-Aware Microservice Deployment for Edge AI Enabled Video Analytics.
WCNC 2024
Zhanpeng Yang, Zhiyong Yu, Xin Liu, Dingzhu Wen, Yong Zhou, and Yuanming Shi

Safe Reinforcement Learning with Instantaneous Constraints: The Role of Aggressive Exploration.
AAAI 2024
Honghao Wei, Xin Liu, and Lei Ying

Sample Efficient Reinforcement Learning in Mixed Systems through Augmented Samples and Its Applications to Queueing Networks.
NeurIPS 2023 (Spotlight)
Honghao Wei, Xin Liu, Weina Wang, and Lei Ying

POBO: Safe and Optimal Resource Management for Cloud Microservices.
Performance 2023
Hengquan Guo, Hongchen Cao, Jingzhu He, Xin Liu, and Yuanming Shi

Federated Linear Bandit Learning via Over-the-air Computation.
GLOBECOM 2023
Jiali Wang, Yuning Jiang, Xin Liu, Ting Wang, and Yuanming Shi

Online Nonstochastic Control with Adversarial and Static Constraints .
ICML 2023
Xin Liu, Zixian Yang, and Lei Ying

Rectified Pessimistic-Optimistic Learning for Stochastic Continuum-armed Bandit with Constraints.
L4DC 2023
Hengquan Guo, Qi Zhu, and Xin Liu

Neural Constrained Combinatorial Bandits.
INFOCOM 2023
Shangshang Wang, Simeng Bian, Xin Liu, and Ziyu Shao

Online Convex Optimization with Hard Constraints: Towards the Best of Two Worlds and Beyond.
NeurIPS 2022
Hengquan Guo, Xin Liu, Honghao Wei, and Lei Ying

Large-System Insensitivity of Zero-Waiting Load Balancing Algorithms.
Sigmetrics 2022
Xin Liu, Kang Gong, and Lei Ying

A Provably-Efficient Model-Free Algorithm for Constrained Markov Decision Processes.
AISTATS 2022 (A Short Version Accepted in ICML 2021 Workshop on Reinforcement Learning Theory)
Honghao Wei, Xin Liu, and Lei Ying

A Provably-Efficient Model-Free Algorithm for Infinite-Horizon Average-Reward Constrained Markov Decision Processes.
AAAI 2022
Honghao Wei, Xin Liu, and Lei Ying

An Efficient Pessimistic-Optimistic Algorithm for Stochastic Linear Bandits with General Constraints.
NeurIPS 2021
Xin Liu, Bin Li, Pengyi Shi, and Lei Ying

Beyond Scaling: Calculable Error Bounds of the Power-of-Two-Choices Mean-Field Model in Heavy-Traffic.
Mobihoc 2021
Hairi, Xin Liu, and Lei Ying

A Simple Steady-State Analysis of Load Balancing Algorithms in the Sub-Halfin-Whitt Regime.
Sigmetrics 2018 MAMA Workshop
Xin Liu and Lei Ying

On Achieving Zero Delay with Power-of-d-choices Load Balancing.
INFOCOM 2018. Fast-Track Review for IEEE TNSE (7 out of 312 accepted papers were invited at INFOCOM 2018)
Xin Liu and Lei Ying

Wireless Scheduling with Deadline and Power Constraints.
CISS 2018
Yiqiu Liu, Xin Liu, Lei Ying, and R. Srikant

Fluid-model-based Car Routing for Modern Ridesharing Systems.
Sigmetrics 2017 (Poster)
Anton Braverman, Jim Dai, Xin Liu, and Lei Ying

Spatial-temporal Routing for Supporting End-to-end Hard Deadlines in Multi-hop Networks.
CISS 2016
Xin Liu and Lei Ying

Probability Constrained Robust Multicast Beamforming in Cognitive Radio Network.
ChinaCom 2013. (Best Student Paper)
Xin Liu, Haoqi Li, and Haibin Wang

Journal Articles

Exploration, Exploitation, and Engagement in Multi-Armed Bandits with Abandonment.
Journal of Machine Learning Research, 2024
Zixian Yang, Xin Liu, and Lei Ying

Large-System Insensitivity of Zero-Waiting Load Balancing Algorithms.
Stochastic Systems, 2024
Xin Liu, Kang Gong, and Lei Ying

A Reinforcement Learning and Prediction-Based Lookahead Policy for Vehicle Repositioning in Online Ride-Hailing Systems.
IEEE Transactions on Intelligent Transportation Systems, 2023
Honghao Wei, Zixian Yang, Xin Liu, Zhiwei (Tony) Qin, Xiaocheng Tang, Lei Ying

Federated Reinforcement Learning for Electric Vehicles Charging Control on Distribution Networks.
IEEE Internet of Things Journal, 2023
Junkai Qian, Yuning Jiang, Xin Liu, Qiong Wang, Ting Wang, Yuanming Shi, Wei Chen

Universal Scaling of Distributed Queues Under Load Balancing in the Super-Halfin-Whitt Regime.
IEEE/ACM Transactions on Networking, 2022
Xin Liu and Lei Ying

Steady-State Analysis of Load Balancing with Coxian-2 Distribution Service Times.
Naval Research Logistics, Mar., 2021
Xin Liu, Kang Gong, and Lei Ying

Wireless Scheduling with Deadline and Power Constraints.
Performance Evaluation, Mar., 2021
Yiqiu Liu, Xin Liu, Lei Ying, and R. Srikant

Steady-State Analysis of Load Balancing Algorithms in the Sub-Halfin-Whitt Regime.
Journal of Applied Probability, Apr., 2020
Xin Liu and Lei Ying

On Achieving Zero Delay with Power-of-d-Choices Load Balancing.
IEEE Transactions on Network Science and Engineering. Oct., 2019
Xin Liu and Lei Ying

Empty-Car Routing in Ridesharing Systems.
Operations Research, Aug., 2019. Media coverage: [TechXplore] [Informs Press]
Anton Braverman, Jim Dai, Xin Liu, and Lei Ying

Spatial-Temporal Routing for Supporting End-to-End Hard Deadlines in Multi-Hop Networks.
Performance Evaluation, July, 2019
Xin Liu, Weichang Wang, and Lei Ying

Joint Beamforming and User Selection in Multicast Downlink Channel under Secrecy-outage Constraint.
IEEE Communications Letters , Jan., 2014
Xin Liu, Feifei Gao, Gongpu Wang, and Xiyuan Wang

Preprints

Adversarially Trained Actor Critic for offline CMDPs.
Submitted
Honghao Wei, Xiyue Peng, Xin Liu, Arnob Ghosh

Scalable and Sample Efficient Distributed Policy Gradient Algorithms in Multi-Agent Networked Systems.
Xin Liu, Honghao Wei, and Lei Ying

POND: Pessimistic-Optimistic oNline Dispatching.
A Short Version Accepted in Sigmetrics 2021 RLNQ Workshop
Xin Liu, Bin Li, Pengyi Shi, and Lei Ying

Group Members

- Hengquan Guo (PhD, 2021)
- Qi Zhu (Master, 2022)
- Botao Ye (Master, 2022)
- Yongxin Xu (Master, 2022)
- Xinrui Liu (Master, 2023)
- Haobo Zhang (Master, 2023)
- Xiyue Peng (Master, 2024)
Professional Service

Program Committee: INFOCOM 2022~2024, MOBIHOC 2021~2024, Performance 2023, ICDCS 2023~2024,
ITC 33~35, WiOpt 2021, 2024.
Reviewer: IEEE/ACM Transactions on Networking, Performance Evaluation, IEEE Journal on Selected Areas in Communications, IEEE Transactions on Information Theory, ICML, NeurIPS, AISTATS, INFOCOM, MOBIHOC, ICDCS.


Credit to Jon