Selected Publications

This page contains selected recent publications available in a wide array of formats and organized by the research topics. This research was supported in part by various funding agencies of both China and USA governments and industries in the past. The active and recently completed research projects which I’m leading include MOST of China Key Project on Smart City (科技部智慧城市专项) under grant 2019YFB2102100 (12/2019-11/2022), Guangdong Key R&D Project on cloud computing under grant 2020B010164003 (1/2020-12/2022), FDCT of Macao Key R&D Program on AI under grant 0015/2019/AKP (1/2020-12/2022), 深圳市城市计算与数据智能学科建设(2017-2020) 及深圳北斗位置服务技术工程研究中心(I/II期,1/2013-12/2021), and GD-HK-MO Joint Lab of Human-Machine Intelligence Synergy Systems (粵港澳人機智能協同系統聯合實驗室)

News: congrats to the team on recent great work in

Distributed and Cloud Systems: ISCA’2024, ACM TOCS’2023, Supercomputing’2023, ASPLOS’2023, Eurosys’2023,  HPDC’2023, ICDCS’2023, MICRO’2022, SoCC’2022,  ACM middleware’2022,  HPDC’2022, ICDCS’2022, ACM SoCC’2021(best paper award), INFOCOM’2021 (2 papers), RTSS’2019. IPDPS’2019, ICDCS’2019

AI Theory and Apps: AAAI’2024(6), CVPR’2023 (3), ICDE’2023, NeurIPS’2022, ECCV’2022 (2), CVPR’2022, AAAI’2022,  ACM sigspatial’2022, CVPR ‘2021 (1 oral + 2 regular), ICCV’2021, ICLR’2020, ICML’2020, ICLR’2019, NeurIPS’2019

Overview of Research Outputs:

  • See AI-overview for achievements of AI fundamentals in robust deep learning, transfer learning, federated learning, and reinforcement learning.
  • See  DIUT-summary.ppt  and FC-UPS.ppt for activities and achievements on Data Intelligence for Urban Transportation and Smart Cities.
  • Talk on Effective Computing for AI covers part of the results about cloud and cloud-edge infrastructures for AI applications.
  • Talk on Connected Autonomous Driving summarizes research results in MoCAD project

Click the following links for representative papers in different topics.  Please visit Google Scholar  or  DBLP library for list of publications, or scan the following QR codes.  ORCID: 000-001-9480-0356, Scopus ID: 55600419500

  • Optimizing Resource Management for Shared Microservices: A Scalable System Design, ACM Transactions on Computer Systems, 2023
  • Interference-aware Multiplexing for Deep Learning in GPU Clusters: A Middleware Approach, SC’2023
  • The power of prediction: microservice auto scaling via workload learning, SoCC’2022
  • HARMONY: Heterogeneity-Aware Hierarchical Management for Federated Learning System, IEEE/ACM Int’l Conf. on Microarchitecture, 2022
  • Erms: Efficient Resource Management for Shared Microservices with SLA Guarantees, ACM ASPLOS’2023
  • PDMA: Probabilistic service migration approach for delay-aware and mobility-aware mobile edge computing. Softw. Pract. Exp. 52(2): 394-414 (2022)
  • Improving Concurrent GC for Latency Critical Services in Multi-tenant Systems, ACM/IFIP Middlware‘ 2022
  • Characterizing Microservice Dependency and Performance: Alibaba Trace Analysis, ACM SoCC’2021 (best paper award)
  • Cost-Driven Data Caching in the Cloud: An Algorithmic Approach, IEEE Infocom 2021.
  • SmartDistance: A Mobile-based Positioning System for Automatically Monitoring Social Distance, IEEE Infocom 2021
  • Multi-layer Coordination for High-Performance Energy-Efficient Federated Learning. IWQoS 2020: 1-10
  • RS-pCloud: An Peer-to-Peer Based Edge-Cloud System for Fast Remote Sensing Image Processing, IEEE EDGE 2020 (best student paper award)
  • SmartPC: Hierarchical Pace Control in Real-Time Federated Learning System. IEEE RTSS 2019: 406-418
  • N. Tziritas, et al. Online live VM migration algorithms to minimize total migration time and downtime, IEEE IPDPS’2019.
  • G. Xu and C. Xu, MEER: online estimation of optimal memory reservations for long lived containers in in-memory cluster computing. IEEE ICDCS’2019.
  • Y. Zhao*, X. Gao*(equal), et al. Mayo: A Framework for Auto-generating Hardware Friendly Deep Neural Networks. In MobiSys EMDL workshop, Jun 2018.
  • R. Li, C. Shen, H. He, Z. Xu and C. Xu. A lightweight secure data sharing scheme for mobile cloud computing. IEEE Transactions on Cloud Computing, April 2018.
  • N. Tziritas, et al, Data replication and virtual machine migrations to mitigate network overhead in edge computing systems. IEEE Transactions on Sustainable Computing, June 2017.
  • S. He, et al, Heterogeneity-aware collective I/O for parallel I/O systems with hybrid HDD/SSD servers, IEEE Transactions on Computers, 66(6):1091-1098, 2017.
  • S. He, et al, HARL: optimizing parallel file systems with heterogeneity-aware region-level data layout. IEEE Transactions on Computers, 66(6):1048-1060, June 2017.
  • Y. Wang, et al. On service migrations in the cloud for mobile accesses: a distributed approach, ACM Transactions on Autonomous Adaptive Systems, 12(2):1-25, May 2017.
  • L. Zeng, et al, Raccoon: a novel network I/O allocation framework for workload-aware VM scheduling in virtual environments. IEEE Trans. on Parallel and Distributed Systems, 28(9):2651-2662, Sept 2017.
  • S. He, et al, Using minmax-memory claims to improve in-memory workflow computations in the cloud, IEEE Trans. on Parallel and Distributed Systems, 28(4):1202-1214, April 2017.
  • T. Maqsood, et al, Leveraging on deep memory hierarchies to minimize energy consumption and data access latency on single-chip cloud computers. IEEE Transactions on Sustainable Computing, 2(2):154-166, February 2017.
  • L. Zeng, S. Xu, Y. Wang, K. Kent, D. Bremner, C. Xu, Toward cost-effective replica placements in cloud storage systems with QoS-awareness. Software: Practice and Experience, 47(6):813-829, June 2017.
  • Y. Liu, et al, Barrier-Aware warp scheduling for throughput processor, ACM Int. Conf. on Supercomputing (ICS), June 2016.
  • G. Xu, C. Xu and S. Jiang. Prophet: Scheduling Executors with Time-Varying Resource Demands on Data-Parallel Computation Frameworks. Proc. ICAC 2016: 45-54.
  • D. Dilli, et al, A low disk-bound transaction logging system for in-memory distributed data stores, IEEE Proc. of Cluster Computing , September 2016 (Best paper award nominee).
  • N. Tziritas, et al, On improving constrained single and group operator placement using evictions in big data environments, IEEE Transactions on Service Computing, Sept 2016.
  • Z. Bei, et al, RFHOC: A random-forest approach to auto-tuning Hadoop’s configuration. IEEE Trans. on Parallel and Distributed Systems, 27(5): 1470-1483, May 2016.
  • C. Jiang, et al, Two-Level Hybrid Sampled Simulation of Multithreaded Applications. ACM Transactions on Architecture and Code Optimization (TACO), 12(4): 39:1-39:25, April 2016.
  • BAT: Behavior-Aware Human-Like Trajectory Prediction for Autonomous Driving, AAAI’2024
  • HCPerf: Driving Performance-Directed Hierarchical Coordination for Autonomous Vehicles, ICDCS’2023
  • Weakly Supervised Monocular 3D Object Detection Using Multi-View Projection and Direction Consistency, CVPR’2023
  • Context‐aware trajectory prediction for autonomous driving in heterogeneous environments, Computer‐Aided Civil and Infrastructure Engineering, 2023
  • Distributed Data-Sharing Consensus in Cooperative Perception of Autonomous Vehicles, ICDCS’2022
  • MORA: Improving ensemble robustness evaluation with model-reweighing attack, NeurIPS’2022
  • Boosting Active Learning via Improving Test Performance. AAAI 2022
  • Semi-supervised 3D Object Detection with Proficient Teachers. ECCV (38) 2022
  • HiTPR: Hierarchical Transformer for Place Recognition in Point Cloud,ICRA’2022
  • Ada-Detector: Adaptive Frontier Detector for Rapid Exploration, ICRA’2022
  • Grounding Commands for Autonomous Vehicles via Layer Fusion with Region-specific Dynamic Layer Attention. IROS 2022
  • Learning Moving-Object Tracking with FMCW LiDAR. IROS 2022
  • Peer-Assisted Robotic Learning: A Data-Driven Collaborative Learning Approach for Cloud Robotic Systems, ICRA’2021
  • A general elimination strategy for camera motion estimation, ICRA’2021
  • LAFEAT: Piercing Through Adversarial Defenses with Latent Features, CVPR’2021 (oral presentation) [Robustness and Security]
  • Focused Quantization for Sparse CNNs. NeurIPS 2019: 5585-5594 [Model inference efficiency in AD]
  • Dynamic channel pruning: feature boosting and suppression, ICLR 2019 [Model inference efficiency in AD]
  • RIFLE: Backpropagation in Depth for Deep Transfer Learning through Re-Initializing the Fully-connected LayEr, ICML‘2020, [Model buildup in small data samples]
  • FPGA Implementation for CNN acceleration,  FTS’2019 [HW/SW co-design for model inference]
  • B. Y. Liu, L. J. Wang, X. Q. Chen, L. X. Huang, C. Z. Xu, “Peer-Assisted Robotic Learning: A Data-Driven Collaborative Learning Approach for Cloud Robotic Systems”, in Proc. ICRA 2021.
  • Federated Imitation Learning: A Novel Framework for Cloud Robotic Systems With Heterogeneous Sensor Data. IEEE Robotics Autom. Lett. 5(2): 3509-3516 (2020)
  • FC-SLAM: Federated Learning Enhanced Distributed Visual-LiDAR SLAM In Cloud Robotic System. ROBIO 2019
  • Boyi LiuLujia WangMing LiuChengzhong Xu: Lifelong Federated Reinforcement Learning: A Learning Architecture for Navigation in Cloud Robotic Systems. CoRR abs/1901.06455 (2019)
  • A Robust Stereo Camera Localization Method with Prior LiDAR Map Constrains. ROBIO 2019
  • TrafficAdaptor: An adaptive obfuscation strategy for vehicle location rrivacy against traffic flow aware attacks, ACM SIGSPATIAL, 2022
  • Li Li, et al, SmartDistance:A mobile based positioning system for automatically monitoring social distanceINFOCOM 2021: 1-10 (PDF
  • L. Yan, et al, Employing opportunistic charging for electric taxicabs to reduce idle time, Proc of ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT), vol 2(1), 2018 (Proc. of ACM on Ubicomp, 2018).
  • H. Zhang, et al, Urban-Scale Human Mobility Modeling With Multi-Source Urban Network Data. IEEE/ACM Transactions on Networking 26(2): 671-684 (2018).
  • Z. Yu, et al, MIA: Metric importance analysis for big data workload characterization, IEEE Trans. on Parallel and Distributed Systems, 29(6):1371-1384, June 2018.
  • L. Yan, et al. CatCharger: Deploying wireless charging lanes in a metropolitan road network through categorization and clustering of vehicle traffic. INFOCOM 2017:1-9.
  • B. Zhao, et al, A data-driven congestion diffusion model for characterizing traffic in metrocity scales, Proc. of IEEE Conf. on Big Data, 2017.
  • FFD: A Federated Learning Based Method for Credit Card Fraud Detection. BigData Congress 2019 (PDF)
  • J. Zhao, et al, Estimation of passenger route choice pattern using smart card data for complex metro systems. IEEE Trans. on Intelligent Transportation Systems, 18(4):790-801, April 2017.
  • Y. Zhao, et al, ER-CRLB: An Extended Recursive Cramér-Rao Lower Bound Fundamental Analysis Method for Indoor Localization Systems, IEEE Transactions on Vehicular Technology, 66(2):1605-1618, February 2017.
  • F. Zhang, et al. Spatiotemproal segmentation of metro trips using smart card data. IEEE Transactions on Vehicular Technology, 65(3):1137-1149, March 2016.
  • F. Tang, M. Guo, S. Guo, C. Xu. Mobility prediction based joint stable routing and channel assignment for mobile ad hoc cognitive networks. IEEE Trans. on Parallel and Distributed Systems, 27(3): 789-803, March 2016.
  • M. Chen, et al, A novel approach to system design for dialect speech interaction with NAO robot, ICAR 2017:476-481.
  • Z. He, et al. Exploiting Real-Time Traffic Light Scheduling with Taxi Traces. ICPP 2016: 314-323.
  • Z. Tian, et al. Real-time charging station recommendation system for electric-vehicle taxis. IEEE Trans. on Intelligent Transportation Systems, 17(11): 3098-3109, Nov. 2016.
  • W. Xiong, et al. ShenZhen transportation system (SZTS): a novel big data benchmark suite. The Journal of Supercomputing 72(11): 4337-4364, Nov. 2016.

See AI-overview for brief descriptions of  fundamental AI research   

Adversarial Attack Evaluation:

  • LAFIT: Efficient anLAFIT: Efficient and Reliable Evaluation of Adversarial Defenses with Latent Features, TPAMI, 2023
  • MORA: Improving Ensemble Robustness Evaluation with Model Reweighing Attack, NeurIPS 2022
  • Efficient Loss Function by Minimizing the Detrimental Effect of Floating-Point Errors on Gradient-Based Attacks, CVPR’2023
  • LAFEAT: Piercing Through Adversarial Defenses with Latent Features, CVPR’2021 (oral presentation

Transfer Learning:

  • RIFLE: Backpropagation in Depth for Deep Transfer Learning through Re-Initializing the Fully-connected LayEr. ICML 2020: 6010-6019
  • Improving Bert Fine-Tuning via Stabilizing Cross-Layer Mutual Information, ICASSP’2023
  • SMILE: Sample-to-feature Mixup for Efficient Transfer Learning, Transactions on Machine Learning Research, 2023
  • Knowledge Distillation with Attention for Deep Transfer Learning of Convolutional Networks. ACM Trans. Knowl. Discov. Data 16(3), 2022
  • Boosting Active Learning via Improving Test Performance, AAAI 2022
  • Adaptive Consistency Regularization for Semi-Supervised Transfer Learning. CVPR 2021
  • Pay Attention to Features, Transfer Learn Faster CNNs. ICLR 2020

Model Pruning and Compression:

  • Dynamic channel pruning: feature boosting and suppression, ICLR 2019
  • Focused Quantization for Sparse CNNs. NeurIPS 2019: 5585-5594 
  • FPGA Implementation for CNN acceleration,  FTS’2019 [HW/SW co-design for model inference]
  • SpHMC: Spectral Hamiltonian Monte Carlo, AAAI, 2019.

Federated Learning:

  • FedDC: Federated Learning with Non-IID Data via Local Drift Decoupling and Correction, CVPR 2022
  • Adaptive Channel Sparsity for Federated Learning under system heterogeneity, CVPR2023

Reinforcement Learning and Multiagent:

  • Model-free λ-policy iteration for discrete-time linear quadratic regulation, TNNLS’2023
  • Adaptive Fuzzy Leader-Follower Synchronization of Constrained Heterogeneous Multiagent Systems. IEEE Trans. Fuzzy Syst. 30(1): 205-219 (2022)
  • Robust Actor-Critic Learning for Continuous-Time Nonlinear Systems With Unmodeled Dynamics. IEEE Trans. Fuzzy Syst. 30(6): 2101-2112 (2022)

Publications related to Recently Completed Projects