Jingwei Sun

jingwei.sun@duke.edu

Durham, NC, USA

I am currently a Ph.D. candidate in the Electrical and Computer Engineering department at Duke University under the supervision of Professor Yiran Chen and Hai “Helen” Li in Duke CEI Lab. My research interests lie in machine learning systems and edge computing, with a focus on collaborative intelligent systems and on-device AI with the awareness of effciency, security and privacy.

selected publications

arxiv

Knowledge Graph Tuning: Real-time Large Language Model Personalization based on Human Feedback

Jingwei Sun, Zhixu Du, and Yiran Chen

arXiv preprint arXiv:2405.19686, 2024

PDF

ICML

FedBPT: Efficient Federated Black-box Prompt Tuning for Large Language Models

Best Paper Award in Federated Learning on the Edge, 2024 AAAI Spring Series Symposium

Jingwei Sun, Ziyue Xu, Hongxu Yin, and 6 more authors

Proceedings of the 41th International Conference on Machine Learning, 2024

Bib PDF Website

@article{sun2023fedbpt,
  title = {FedBPT: Efficient Federated Black-box Prompt Tuning for Large Language Models},
  author = {Sun, Jingwei and Xu, Ziyue and Yin, Hongxu and Yang, Dong and Xu, Daguang and Liu, Yudong and Du, Zhixu and Chen, Yiran and Roth, Holger R},
  journal = {Proceedings of the 41th International Conference on Machine Learning},
  year = {2024},
  award = {Best Paper Award in Federated Learning on the Edge, 2024 AAAI Spring Series Symposium},
}

ICCV

Communication-Efficient Vertical Federated Learning with Limited Overlapping Samples

Jingwei Sun, Ziyue Xu, Dong Yang, and 6 more authors

2023

Bib PDF

@article{sun2023communication,
  title = {Communication-Efficient Vertical Federated Learning with Limited Overlapping Samples},
  author = {Sun, Jingwei and Xu, Ziyue and Yang, Dong and Nath, Vishwesh and Li, Wenqi and Zhao, Can and Xu, Daguang and Chen, Yiran and Roth, Holger R},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision},
  year = {2023},
}

SenSys
FedSEA: A Semi-Asynchronous Federated Learning Framework for Extremely Heterogeneous Devices

Jingwei Sun, Ang Li, Lin Duan, and 8 more authors

2022

Abs Bib HTML PDF

Federated learning (FL) has attracted increasing attention as a promising technique to drive a vast number of edge devices with artificial intelligence. However, it is very challenging to guarantee the efficiency of a FL system in practice due to the heterogeneous computation resources on different devices. To improve the efficiency of FL systems in the real world, asynchronous FL (AFL) and semi-asynchronous FL (SAFL) methods are proposed such that the server does not need to wait for stragglers. However, existing AFL and SAFL systems suffer from poor accuracy and low efficiency in realistic settings where the data is non-IID distributed across devices and the on-device resources are extremely heterogeneous. In this work, we propose FedSEA - a semi-asynchronous FL framework for extremely heterogeneous devices. We theoretically disclose that the unbalanced aggregation frequency is a root cause of accuracy drop in SAFL. Based on this analysis, we design a training configuration scheduler to balance the aggregation frequency of devices such that the accuracy can be improved. To improve the efficiency of the system in realistic settings where the devices have dynamic on-device resource availability, we design a scheduler that can efficiently predict the arriving time of local updates from devices and adjust the synchronization time point according to the devices’ predicted arriving time. We also consider the extremely heterogeneous settings where there exist extremely lagging devices that take hundreds of times as long as the training time of the other devices. In the real world, there might be even some extreme stragglers which are not capable of training the global model. To enable these devices to join in training without impairing the systematic efficiency, Fed-SEA enables these extreme stragglers to conduct local training on much smaller models. Our experiments show that compared with status quo approaches, FedSEA improves the inference accuracy by 44.34% and reduces the systematic time cost and local training time cost by 87.02X and 792.9X. FedSEA also reduces the energy consumption of the devices with extremely limited resources by 752.9X.
@article{sun2022fedsea, title = {FedSEA: A Semi-Asynchronous Federated Learning Framework for Extremely Heterogeneous Devices}, author = {Sun, Jingwei and Li, Ang and Duan, Lin and Alam, Samiul and Deng, Xuliang and Guo, Xin and Wang, Haiming and Gorlatova, Maria and Zhang, Mi and Li, Hai and others}, booktitle = {Proceedings of the 20th ACM Conference on Embedded Networked Sensor Systems}, year = {2022}, }

CVPR

Soteria: Provable defense against privacy leakage in federated learning from representation perspective

Jingwei Sun, Ang Li, Binghui Wang, and 3 more authors

In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021

Bib PDF Code

@inproceedings{sun2021soteria,
  title = {Soteria: Provable defense against privacy leakage in federated learning from representation perspective},
  author = {Sun, Jingwei and Li, Ang and Wang, Binghui and Yang, Huanrui and Li, Hai and Chen, Yiran},
  booktitle = {Proceedings of the IEEE/CVF conference on computer vision and pattern recognition},
  pages = {9311--9319},
  year = {2021},
}

NeurIPS

Fl-wbc: Enhancing robustness against model poisoning attacks in federated learning from a client perspective

Jingwei Sun, Ang Li, Louis DiValentin, and 3 more authors

Advances in Neural Information Processing Systems, 2021

Bib PDF Code

@article{sun2021fl,
  title = {Fl-wbc: Enhancing robustness against model poisoning attacks in federated learning from a client perspective},
  author = {Sun, Jingwei and Li, Ang and DiValentin, Louis and Hassanzadeh, Amin and Chen, Yiran and Li, Hai},
  journal = {Advances in Neural Information Processing Systems},
  volume = {34},
  pages = {12613--12624},
  year = {2021},
}

SenSys

Fedmask: Joint computation and communication-efficient personalized federated learning via heterogeneous masking

Ang Li, Jingwei Sun, Xiao Zeng, and 3 more authors

In Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems, 2021

Bib HTML PDF

@inproceedings{li2021fedmask,
  title = {Fedmask: Joint computation and communication-efficient personalized federated learning via heterogeneous masking},
  author = {Li, Ang and Sun, Jingwei and Zeng, Xiao and Zhang, Mi and Li, Hai and Chen, Yiran},
  booktitle = {Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems},
  pages = {42--55},
  year = {2021},
}