Ting-Wu Rudy Chin (金廷武)

alt text 

Ph.D. Candidate,
Electrical and Computer Engineering,
Carnegie Mellon University

E-mail: tingwuc [AT] cmu [DOT] edu

Affiliation: [EnyAC] [OPAL]

Google Scholar

About me

I am a fourth-year Ph.D. student co-advised by Prof. Diana Marculescu and Prof. Gauri Joshi. My research interests include Model Compression, Neural Architecture Search and Trasnsfer Learning for Deep ConvNets.

Before joining CMU, I received the B.S. and M.S. degrees in computer science from National Chiao Tung University (NCTU), in 2015 and 2017, respectively.

Industrial Experiences

alt text 
Facebook AI Research

2020.05 - 2020.08
Research Intern
Mentors: Ari Morcos
Project: On the Transferability of Channel Optimization

alt text 
Facebook Reality Labs

2019.05 - 2019.08
Research Intern
Mentors: Pierce Chuang and Vikas Chandra
Project: One Weight Bitwidth to Rule Them All

alt text 
Microsoft Research

2018.05 - 2018.08
Research Intern
Mentors: Cha Zhang
Project: Efficient Model Compression with Learned Global Ranking

First-authored Peer-reviewed Publications

  • One Weight Bitwidth to Rule Them All
    Ting-Wu Chin, Pierce Chuang, Vikas Chandra, Diana Marculescu
    European Conference on Computer Vision Workshops (ECCV’20 Embedded Vision Workshop) (Best Paper Awards)

  • Towards Efficient Model Compression via Learned Global Ranking
    Ting-Wu Chin, Ruizhou Ding, Cha Zhang, Diana Marculescu
    The IEEE Conference on Computer Vision and Pattern Recognition (CVPR’20) (Oral, 5% Acceptance Rate)

  • AdaScale: Towards Real-time Video Object Detection using Adaptive Scaling
    Ting-Wu Chin, Ruizhou Ding, Diana Marculescu
    Conference on Machine Learning and Systems (MLSys’19) (Oral, 17% Acceptance Rate)

  • Domain-Specific Approximation for Object Detection
    Ting-Wu Chin, Chia-Lin Yu, Matthew Halpern, Hasan Genc, Shiao-Li Tsao, Vijay Janapa Reddi
    IEEE Micro, SI: Autonomous Computing

First-authored Pre-prints and Non-archival Venues

  • PareCO: Pareto-aware Channel Optimization for Slimmable Neural Networks
    Ting-Wu Chin, Ari S. Morcos, Diana Marculescu
    Abridged (4 pages) versions were accepted at KDD 2020 Workshops AdvML and DLP & ICML 2020 Workshops DMMLSys and RealML
    [arXiv] [Workshop PDF] [Code]

  • Improving the Adversarial Robustness of Transfer Learning via Noisy Feature Distillation
    Ting-Wu Chin, Cha Zhang, Diana Marculescu
    An abridged (4 pages) version was accepted at KDD 2020 Workshop AdvML
    [arXiv] [Workshop PDF] [Code]

  • Layer-compensated Pruning for Resource-constrained Convolutional Neural Networks
    Ting-Wu Chin, Cha Zhang, Diana Marculescu
    An abridged (4 pages) version was accepted at NeurIPS 2018 Workshop MLPCD 2 (Oral)
    It is later improved and become our CVPR work “LeGR”

Co-authored Peer-reviewed Publications

  • Regularizing Activation Distribution for Training Binarized Deep Networks
    Ruizhou Ding, Ting-Wu Chin, Diana Marculescu, Zeye Liu
    The IEEE Conference on Computer Vision and Pattern Recognition (CVPR’19)

  • FLightNNs: Lightweight Quantized Deep Neural Networks for Fast and Accurate Inference
    Ruizhou Ding, Zeye Liu, Ting-Wu Chin, Diana Marculescu, R. D. (Shawn) Blanton
    ACM/IEEE Design Automation Conference (DAC’19)

  • Understanding the Impact of Label Granularity on CNN-based Image Classification
    Zhuo Chen, Ruizhou Ding, Ting-Wu Chin, Diana Marculescu
    ICDM 2018 Workshop on Data Science and Big Data Analytics (DSDA)

  • Designing Adaptive Neural Networks for Energy-Constrained Image Classification
    Dimitrios Stamoulis, Ting-Wu Chin, Anand Krishnan Prakash, Haocheng Fang, Sribhuvan Sajja, Mitchell Bognar, Diana Marculescu
    Proceedings of the 37th International Conference on Computer-Aided Design (ICCAD’18)

Honors and Awards

  • Qualcomm Innovation Fellowship 2020 Finalist

  • David Barakat and LaVerne Owen-Barakat Fellowship, Carnegie Institute of Technology

  • 3rd Place for Siemens FutureMakers Challenge at CMU


Reviewer for NeurIPS’20, ICML’20, IEEE Journal of Selected Topics in Signal Processing, NeurIPS’18 Workshop CDNNRIA
Sub-reviewer for MLSys’20