Welcome to my homepage!

Yawen Wu, Ph.D.

E-mail: yawen.wu@pitt.edu

Address: San Diego, California, USA

Home Page, Google Scholar, LinkedIn

I am Yawen Wu, a Senior AI Research Engineer at Qualcomm. I am building high-performance artificial intelligence (AI) technologies while considering resource requirements. Prior to Qualcomm, I was a Postdoctoral Research Associate at the University of Notre Dame working with Dr. Yiyu Shi. Prior to Notre Dame, I obtained my Ph.D. from the University of Pittsburgh advised by Dr. Jingtong Hu. I am open to various kinds of collaboration, please drop me an email if you are interested in my research. 

I received the MICCAI Young Scientist Award Nomination for my work on medical applications of on-device AI. My research internships at Meta and Baidu USA have resulted in technologies delivered to industrial production.

Recent News

Research Interests

Machine Learning, Deep Learning, Large Language Model (LLM), Computer Vision, AI for Healthcare, Self-supervised Learning, Federated Learning, AI Fairness, On-device AI.

Professional Experience

Senior AI Research Engineer - Qualcomm Technologies Inc, San Diego, CA, USA. (May 2023 - Current)

Postdoctoral Research Associate - University of Notre Dame, Notre Dame, IN, USA. (Mar. 2023 – May 2023)

Machine Learning Research Intern - Meta (formerly Facebook), Reality Labs Research, Redmond, WA, USA. (May 2022 - Aug. 2022)

Machine Learning Research Intern, Baidu USA, Sunnyvale, CA, USA. (Sept. 2021 - Dec. 2021)

Selected Publications

Synthetic Data Can Also Teach: Synthesizing Effective Data for Unsupervised Visual Representation Learning (Acceptance rate 19.6%) [pdf]

Yawen Wu, Zhepeng Wang, Dewen Zeng, Yiyu Shi, Jingtong Hu

in Proc. of the Thirty-Seventh AAAI Conference on Artificial Intelligence(AAAI 2023), Feb. 2023.

Additional Positive Enables Better Representation Learning for Medical Images (Early acceptance, acceptance rate 14%) [pdf]

Dewen Zeng, Yawen Wu, Xinrong Hu, Xiaowei Xu, Jingtong Hu, and Yiyu Shi

in Proc. of the 26th Medical Image Computing and Computer Assisted Interventions (MICCAI 2022), Oct. 2023.

Decentralized Unsupervised Learning of Visual Representations (Acceptance rate 15%) [pdf]

Yawen Wu, Zhepeng Wang, Dewen Zeng, Meng Li, Yiyu Shi, Jingtong Hu

in Proc. of the 31st International Joint Conference on Artificial Intelligence(IJCAI 2022), July 2022.

Enabling On-Device CNN Training by Self-Supervised Instance Filtering and Error Map Pruning [Video][Slides][arXiv]

Yawen Wu, Zhepeng Wang, Yiyu Shi, Jingtong Hu

in Proc. of International Conference on Compilers, Architecture, and Synthesis for Embedded Systems (CASES), in conjunction with ESWEEK, Oct. 2020.

Also appears as part of the ESWEEK-TCAD Special Issue, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (IEEE TCAD).

Intermittent Inference with Nonuniformly Compressed Multi-Exit Neural Network for Energy Harvesting Powered Devices [Video][Slides][arXiv]

Yawen Wu, Zhepeng Wang, Zhenge Jia, Yiyu Shi, Jingtong Hu

in Proc. of IEEE/ACM Design Automation Conference (DAC), 2020. (acceptance rate 23%)

Enabling On-Device Self-Supervised Contrastive Learning With Selective Data Contrast [Video][Slides][arXiv]

Yawen Wu, Zhepeng Wang, Dewen Zeng, Yiyu Shi, Jingtong Hu

in Proc. of IEEE/ACM Design Automation Conference (DAC), 2021.

FairPrune: Achieving Fairness Through Pruning for Dermatological Disease Diagnosis (Early acceptance, acceptance rate 13%) [arxiv]

Yawen Wu*, Dewen Zeng* (* equal contribution), Xiaowei Xu, Yiyu Shi, Jingtong Hu

in Proc. of the 25th Medical Image Computing and Computer Assisted Interventions (MICCAI 2022), Sept. 2022.

Federated Contrastive Learning for Dermatological Disease Diagnosis via On-device Learning [arXiv]

Yawen Wu, Dewen Zeng, Zhepeng Wang, Yi Sheng, Lei Yang, Alaina J. James, Yiyu Shi, Jingtong Hu

in Proc. of IEEE/ACM International Conference on Computer-Aided Design (ICCAD 2021), Nov. 2021.

Federated Contrastive Learning for Volumetric Medical Image Segmentation (Early accept, acceptance rate 13%) [arXiv]

Award Nomination: MICCAI Society Young Scientist

Yawen Wu, Dewen Zeng, Zhepeng Wang, Yiyu Shi, Jingtong Hu

in Proc. The 24th Medical Image Computing and Computer Assisted Interventions (MICCAI 2021), Sept. 2021.

Distributed Contrastive Learning for Medical Image Segmentation [pdf]

Yawen Wu, Dewen Zeng, Zhepeng Wang, Yiyu Shi, Jingtong Hu

Medical Image Analysis (in print) (Impact factor=8.5), 2022.

Cooperative Communication Between Two Transiently Powered Sensor Nodes by Reinforcement Learning

Yawen Wu, Zhenge Jia, Fei Fang, Jingtong Hu

IEEE Transactions on COMPUTER-AIDED DESIGN of Integrated Circuits and Systems (TCAD), Jan. 2021.

Positional Contrastive Learning for Volumetric Medical Image Segmentation [arXiv]

Dewen Zeng, Yawen Wu, Xinrong Hu, Xiaowei Xu, Haiyun Yuan, Meiping Huang, Jian Zhuang, Yiyu Shi, Jingtong Hu

in Proc. The 24th Medical Image Computing and Computer Assisted Interventions (MICCAI 2021), Sept. 2021.

The Larger The Fairer? Small Neural Networks Can Achieve Fairness for Edge Devices

Yi Sheng, Junhuan Yang, Yawen Wu, Kevin Mao, Yiyu Shi, Jingtong Hu, Weiwen Jiang, Lei Yang

in Proc. The 59th IEEE/ACM Design Automation Conference (DAC 2022) , July 2022.

Energy-Aware Adaptive Multi-Exit Neural Network Inference Implementation for a Millimeter-Scale Sensing System

Yuyang Li, Yawen Wu, Xincheng Zhang, Jingtong Hu, Inhee Lee

IEEE Transactions on Very Large Scale Integration (VLSI) Systems (TVLSI), April 2022.

Algorithm-Hardware Co-design of Attention Mechanism on FPGA Devices

Xinyi Zhang, Yawen Wu, Peipei Zhou, Xulong Tang, Jingtong Hu

in Proc. International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS) in conjunction with (ESWEEK), Virtual, Oct. 2021.

Also appears as part of the ESWEEK-TECS Special Issue, ACM Transactions on Embedded Computing Systems (ACM TECS).

Enabling Weakly-Supervised Temporal Action Localization from On-Device Learning of the Video Stream

Yue Tang, Yawen Wu, Peipei Zhou, Jingtong Hu

in Proc. International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS) in conjunction with (ESWEEK), Shanghai, China, Oct. 7-14, 2022.

Also appears as part of the ESWEEK-TECS Special Issue, ACM Transactions on Embedded Computing Systems (ACM TECS).

Developing a Miniature Energy-Harvesting-Powered Edge Device with Multi-Exit Neural Network

Yuyang Li, Yawen Wu, Xincheng Zhang, Ehab Hamed, Jingtong Hu, Inhee Lee

in Proc. IEEE Int'l Symposium on Circuits & Systems (ISCAS 2021), May, 2021.

Lightweight Run-Time Working Memory Compression for Deployment of Deep Neural Networks on Resource-Constrained MCUs

Zhepeng Wang, Yawen Wu, Zhenge Jia, Yiyu Shi, Jingtong Hu

The 26th Asia and South Pacific Design Automation Conference (ASP-DAC 2021), Jan. 2021.

Cooperative Communication Between Two Transiently Powered Sensors by Reinforcement Learning: Work-in-Progress

Yawen Wu, Zhenge Jia, Fei Fang, Jingtong Hu

in Proc. of International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS). IEEE, 2019.

Prototyping Energy Harvesting Powered Systems with Nonvolatile Processor

Yawen Wu, Yinan Sun, Zhenge Jia, Lefan Zhang, Yongpan Liu, Jingtong Hu

in Proc. of International Symposium on Rapid System Prototyping (RSP). IEEE, 2018.

Implementation of Multi-Exit Neural-Network Inferences for an Image-Based Sensing System with Energy Harvesting

Yuyang Li, Yuxin Gao, Minghe Shao, Joseph T. Tonecha, Yawen Wu, Jingtong Hu, Inhee Lee

Journal of Low Power Electronics and Applications (JLPEA), Sept. 2021.

Opportunistic Communication with Latency Guarantees for Intermittently-Powered Devices

Kacper Wardega, Wenchao Li, Hyoseung Kim, Yawen Wu, Zhenge Jia, Jingtong Hu

in Proc. The ACM/IEEE Design, Automation and Test in Europe (DATE 2022) , ANTWERP, BELGIUM, March 2022.

Professional Services

TPC and Reviewer  

Teaching Experience

Supervised Students

Awards