Who i AM
I got my PhD degree from George Mason University in May 2022 (Advisor: Dr. Xiang Chen). My recent researches focus on deep learning on LLM-based RAG, embedded systems, deep learning security. Earlier, I received my Bachelor and Master degrees from the School of Electronic and Information Engineering at Beijing Jiaotong University.
News:
05/2024 One collaborative work was accepted by ICML 2024!
02/2024 Our contributed book chapter will be published in "Federated Learning: Theory and Practice".
11/2023 Our paper received the Best Paper Award Nomination in ASP-DAC 2024!
09/08/2023 One collaborative work on Quadratic DNN is accepted by ASP-DAC 2024.
06/07/2023 One collaborative work on Quadratic DNN is accepted in DAC 2023 WIP.
06/06/2023 I will serve as finance chair of IEEE Cloud Summit 2023.
10/20/2022 I served as a session chair at IEEE Cloud Summit 2022.
09/01/2022 Our work got Best Paper Award in Cross-FL 2022 Workshop (link).
08/28/2022 I attend MLSys 2022 and give a presentation.
08/20/2022 I will serve as TPC in IEEE Cloud Summit 2022.
06/01/2022 I will join CVS Health this summer.
04/11/2022 Our work on GPU-aware DNN design (poster) is accepted in EuroSys'22 !
01/14/2022 Our paper is accepted by MLSys 2022 (Top 5 Outstanding Paper award 1.9%) (link).
01/03/2022 Our paper is accepted by IEEE Transactions on Computer-Aided Design of Intergrated Circuits and Systems.
11/01/2021 I am selected as DAC Young Fellow 2021
10/04/2021 Our paper "FalCon: Fine-grained Feature Map Sparsity Computing with Decomposed Convolutions for Inference Optimization" is accepted by WACV2022.
05/16/2021 Our paper is accepted by KDD.
02/24/2021 Our paper is accepted by the IEEE Transactions on Computer-Aided Design of Intergrated Circuits and Systems.
02/12/2021 Our paper "Helios: Heterogeneity-Aware Federated Learning with Dynamically Balanced Collaboration" is accepted by DAC2021.
12/01/2020 One project (Privacy-preserving FL with personal mobility data) won the first prize in IEEE Services Hackathon 2020 (Link)
05/03/2020 Our paper "DiReCtX: Dynamic Resource-Aware CNN Reconfiguration Framework for Real-Time Mobile Applications " is accepted by the IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.
02/01/2020 I will work as an instructor of ECE 445 LAB in ECE Department for this spring semester.
01/14/2020 I made a presentation in ASP-DAC 2020.
07/14/2019 I made a presentation about face/object detection on edge devices at Comcast Applied AI.
06/20/2019 I presented a camera object detection project at Comcast Lab Week.
06/01/2019 I made a presentation on KDD AIoT 2019. [The presentation is based on one of our current works - "DoPa". The original draft on arXiv [Arxiv] was modified and submitted to a conference already, while a conceptual short abstract was submitted only for this presentation [KDD version].]
02/10/2019 Two of our papers are accepted by DAC 2019.
01/15/2019 I will work as a teaching assistant in ECE Department for this semester.
01/10/2019 I will intern at Comcast Research Labs this summer.
RESEARCH INTERESTS
ML pipeline and autoML
High Performance Mobile Computing System
Deep Learning Model(esp. Deep Neural Networks) Reconfiguration and Compression
Mobile Intelligent Application Robustness and Security (e.g. Automatic Speech Recognition)
Research Projects
High Order Convolutional Neural Network
High-Performance Mobile System for Deep Learning
Explored the resource-aware dynamic DNN reconfiguration for mobile systems.
Mobile Intelligence System Robustness, Security, and Authentication
Explored the robustness enhancement algorithm for deep learning models;
Explored the mobile security enhancement with deep learning-powered applications, such as facial recognition and automatic speech recognition.
Vision Sensors based Autonomous Localization for Underground Vehicles
Built the vision-based 3-D localization algorithm for autonomous vehicles;
Landmark recognition optimization under dim light condition;
Designed the practical evaluation experiment in the real underground tunnel.
Wireless Mesh Network Robustness and Topology
The mesh network structure optimization for better scalability;
Evaluated the robustness of data transmission link protocols.
Publication
2024
[ICML'24] C. Xu, F. Yu, Z. Xu, N Inkawhich, and X. Chen. Out-of-Distribution Detection via Deep Multi-Comprehension Ensemble, in Proceedings of International Conference on Machine Learning 2024, To appear.
[ASP-DAC'24] C. Xu, F. Yu, Z. Xu, C. Liu, J. Xiong, and X. Chen. QuadraNet: Improving High-Order Neural Interaction Efficiency with Hardware-Aware Quadratic Neural Networks, in Proceedings of 29th Asia and South Pacific Design Automation Conference (Best Paper Award Nomination), 2024
[Book Chapter] Xiang Chen, Zirui Xu, and Fuxun Yu. Federated Learning: Theory and Practice. ISBN 9780443190377. 2024
2022
[EuroSys '22 Poster] F. Yu, Z. Xu, T. Shen, D. Stamoulis, L. Shangguan, D. Wang, M. Zhang, X. Tong, R. Madhok, C. Zhao, X. Li, N. Karianakis, D. Lymberopoulos, C. Liu, A. Li, Y. Chen, and X. Chen. Rethinking Latency-aware DNN Design with GPU Tail Effect Analysis, poster accepted in the 17th European Conference on Computer Systems (EuroSys), to appear.
[MLSys '22] Z. Xu, F. Yu, J. Xiong, and X. Chen. QuadraLib: A Performant Quadratic Neural Network Library for Architecture Optimization and Design Exploration, in Proceedings of the Fifth Conference on Machine Learning and Systems (MLSys), to appear (Outstanding Paper Award top 1.9%).
[WWW '22] Y. Yu, F. Yu, Z. Xu, D. Wang, M. Zhang, and X. Chen. Powering Multi-Task Federated Learning with Competitive GPU Resource Sharing, In Companion Proceedings of the Web Conference 2022 , 2022 (short paper).
[T-CAD] F. Yu, Z. Xu, C. Liu, D. Stamoulis, D. Wang, Y. Wang, and X. Chen. AntiDoteX: Attention-based Dynamic Optimization for Neural Network Runtime Efficiency, in the Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), to appear.
[WACV '22] Z. Xu, F. Yu, Z. Wu, H. Wang, and X. Chen. FalCon: Fine-grained Feature Map Sparsity Computing with Decomposed Convolutions for Inference Optimization, in Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV), Jan. 2022.
2021
[KDD '21] F. Yu, W. Zhang, Z. Qin, Z. Xu, D. Wang, C. Liu, Z. Tian, and X. Chen. Fed^2: Feature Aligned Federated Learning, in Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2021.
[DAC '21] Z. Xu, J. Xiong, F. Yu, and X. Chen. Helios: Heterogeneity-Aware Federated Learning with Dynamically Balanced Collaboration. in Proceedings of the 58th Design Automation Conference (DAC), Dec. 2021.
[T-CAD] Z. Qin, F. Yu, Z. Xu, C. Liu, and X. Chen. CaptorX: A Class-Adaptive Convolutional Neural Network Reconfiguration Framework. IEEE Transactions on Computer-aided Design of Integrate Circuits and Systems.
2020
[IBM-AICS ‘20] Z. Xu, J. Xiong, F. Yu, and X. Chen. Efficient Neural Network Implementation with Quadratic Neuron. The 3rd IBM IEEE CAS/EDS AI Compute Symposium, Nov. 2020.
[T-CAD] Z. Xu, F. Yu, Z. Qin, C. Liu, and X. Chen. DiReCtX: Dynamic Resource-Aware CNN Reconfiguration Framework for Real-Time Mobile Applications. IEEE Transactions on Computer-aided Design of Integrate Circuits and Systems.
[ASP-DAC '20] Z. Xu, F. Yu and X. Chen. LANCE: AComprehensive and Lightweight CNN Defense Methodology against Physical Adversarial Attacks on Embedded Multimedia Applications . in Proceedings of the 25th Asia and South Pacific Design Automation Conference, Jan. 2020.
2019
[DAC '19] Z. Xu, F. Yu, C. Liu, and X. Chen. ReForm: Static and Dynamic Resource-Aware DNN Reconfiguration Framework for Mobile Devices. in Proceedings of the 56th Design Automation Conference, Jun. 2019.
[DAC '19] F. Yu*, Z. Xu*, C. Liu, and X. Chen. MASKER: Adaptive Mobile Security Enhancement against Automatic Speech Recognition in Eavesdropping. in Proceedings of the 56th Design Automation Conference, Jun. 2019.
[ASP-DAC '19] Z. Xu, F. Yu, C. Liu, and X. Chen. HAMPER: High-Performance Adaptive Mobile Security Enhancement against Malicious Speech and Image Recognition. in Proceedings of the 24th Asia and South Pacific Design Automation Conference, Jan. 2019.
2018
[ISLPED '18] Z. Xu, Z. Qin, F. Yu, C. Liu, and X. Chen. DiReCt: Resource-Aware Dynamic Model Reconfiguration for Convolutional Neural Network in Mobile Systems. in Proceedings of the 23rd ACM/IEEE International Symposium on Low Power Electronics and Design, No. 37, pp. 1 6, Jul. 2018.
[DAC WIP '18] Z. Xu, F. Yu, and X. Chen. DiReCt: Performance-Aware Dynamic Model Reconfiguration for Convolutional Neural Network in Mobile Systems. in the 55th Design Automation Conference, Working In Progress, Jun. 2018.
[EdgeSP '18] F. Yu, Z. Xu, C. Liu, Y. Wang, and X. Chen. Adge: An ADMM-Based Audio Adversarial Example Generation Method. The 3rd ACM/IEEE Workshop on Security and Privacy in Edge Computing, Oct. 2018.
2017
[ICCAD '17] Z. Qin, Z. Xu, Q. Dong, Y. Chen, and X. Chen. VoCaM: Visualization oriented convolutional neural network acceleration on mobile system. in Proceedings of the 36th International Conference on Computer-Aided Design (ICCAD), Page: 835-840, Nov. 2017.
[ICCAD '17] J. Mao, Z. Qin, Z. Xu, K. Nixon, X. Chen, H. Li, and Y. Chen. AdaLearner: An Adaptive Distributed Mobile Learning System for Neural Networks. in Proceedings of the 36th International Conference on Computer-Aided Design (ICCAD), Page: 835-840, Nov. 2017.
[PLOS ONE] Z. Xu, W. Yang, K. You, W. Li, and Y. Kim. Vehicle Autonomous Localization in Local Area of Coal Mine Tunnel based on Vision Sensors and Ultrasonic Sensors. Plos one, Vol. 12, Jan 2017.
ArXiv
[arXiv '19] Z. Xu, Z. Yang, J. Yang, J. Xiong, and X. Chen. ELFISH: Resource-Aware Federated Learning on Heterogeneous Edge Devices. arXiv:1912.01684, Dec 2019.
[arXiv '20] F. Yu, Z. Xu, and X. Chen. Towards Latency-aware DNN Optimization with GPU Runtime Analysis and Tail Effect Elimination. arXiv:2011.03897, Nov 2020.
[arXiv '20] F. Yu, W. Zhang, Z. Qin, Z. Xu, D. Wang, C. Liu, Z. Tian, and X. Chen. Heterogeneous Federated Learning, Aug. 2020.
[arXiv '18] F. Yu, Z. Xu, Y. Wang, C. Liu, and X. Chen. Towards Robust Training of Neural Networks by Regularizing Adversarial Gradients. arXiv:1805.09370, May. 2018.
[arXiv '18] Z. Xu, F. Yu, X. Chen. HASP: A High-Performance Adaptive Mobile Security Enhancement Against Malicious Speech Recognition. arXiv:1809.01697, Sep. 2018.
Professional Skill
Programming:
Python, Java, C, Linux, Matlab, CUDA
Machine learning
Tensorflow, PyTorch
Others
Android Development, Solidworks, Latex, Office