Associated Sites

    Please check out my new personal website

    Hello! Thanks for visiting my website! The information here has not been regularly updated since May 2022. For the most up to date information please refer to my new personal website https://sites.google.com/view/huanrui-yang. Thanks!

     

    This is Huanrui Yang (杨幻睿), a Postdoctoral Researcher in the EECS department of UC Berkeley and Berkeley AI Research, under the supervision of Prof. Kurt Keutzer. My primary research interest lies in improving the efficiency and robustness of deep neural network models in real world tasks. Before joining Berkeley I obtained Ph.D. in the Department of Electrical and Computer Engineering at Duke University, under the supervision of Prof. Hai Li and Prof. Yiran Chen in the Duke CEI Lab in 2022. I received B.E. in Electronic Engineering from Tsinghua University in 2017. 

    Education

    • PhD, Electrical and Computer Engineering, Duke University, 08/2017 to 05/2022
      • Dissertation Title: Towards Efficient and Robust Deep Neural Network Models
      • Advisor: Prof. Hai Li and Prof. Yiran Chen
      • Overall GPA:  3.97/4
    • BE, Electronic Engineering, Tsinghua University, 09/2013 to 07/2017
      • Diploma Thesis Title: On-chip Trainable Fully Connected Neural Network Accelerator Architecture (Outstanding Diploma Thesis of Tsinghua University)
      • Advisor: Prof. Yongpan Liu
      • Overall GPA:  90/100
    • Visiting Student, ECE, University of Alberta, 09/2015 to 12/2015
      • Overall GPA:  4.0/4
    • High School, Experimental Class for Gifted Children, Beijing No.8 Middle School, 09/2009 to 06/2013
      • Outstanding Graduates of Beijing No.8 Middle School (Top 10)

    Research Interest

    • Building efficient and robust deep neural networks for real world applications;
    • Mitigate privacy and security issues in crowd-sourcing cloud-based learning system

    Selected Publications

    • Yang, H., Yang, X., Gong, N. Z., & Chen, Y. (2021). HERO: Hessian-Enhanced Robust Optimization for Unifying and Improving Generalization and Quantization Performance. arXiv preprint arXiv:2111.11986. (To be appeared in DAC 2022)      (Paper)
    • Yang, H., Duan, L., & Li, H. (2021). BSQ: Exploring Bit-Level Sparsity for Mixed-Precision Neural Network Quantization. In International Conference on Learning Representations.      (Paper, Code)
    • Yang, H., Zhang, J., Dong, H., … & Li, H. (2020). DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensembles. In Advances in neural information processing systems. (Oral)      (Paper, Code)
    • Li, A., Duan, Y., Yang, H., Chen, Y., & Yang, J. (2020). TIPRDC: Task-Independent Privacy-Respecting Data Crowdsourcing Framework for Deep Learning with Anonymized Intermediate Representations. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. (Best student paper)      (Paper, Code)
    • Yang, H., Tang, M., Wen, W., Yan, F., … & Chen, Y. (2020). Learning Low-rank Deep Neural Networks via Singular Vector Orthogonality Regularization and Singular Value Sparsification. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops.      (Paper, Code)
    • Yang, H., Wen, W., & Li, H. (2020). DeepHoyer: Learning Sparser Neural Network with Differentiable Scale-Invariant Sparsity Measures. In International Conference on Learning Representations.      (Paper, Code)

    Full publication list can be found here.

    Awards and Honors

    Working Experience

    • NVIDIA Corporation, Research Intern, remote, 02/2021-09/2021
    • Microsoft Corporation, Research Intern, Redmond WA, 05/2018-08/2018

    Teaching Experience

    • Leading teaching assistant of ECE 590 / ECE 661 Computer Engineering Machine Learning and Deep Neural Nets (Fall 2019, Fall 2020 & Fall 2021), Duke University
    • Teaching assistant of ECE 681 Pattern Classification and Recognition (Spring 2019), Duke University
    • Teaching assistant of ECE 550D Fundamentals of Computer Systems and Engineering (Fall 2018), Duke University