Lei Hsiung

Visiting Scholar @ IBM Research  •  Previously: NTHU / MediaTek / Logitech

leihsiung.ray [at] gmail.com

:sparkles: I am looking for a Ph.D. position starting in Fall 2023! :sparkles:

About me: I am a research assistant working with Prof. Tsung-Yi Ho at National Tsing Hua University (NTHU). Currently, I visit IBM Thomas J. Watson Research Center in New York and do machine learning research with Dr. Pin-Yu Chen. Previously, I completed my MSc in Computer Science at NTHU in 2022; I also got my BSc in Mathematics and Computer Science from NTHU in 2020. I had also spent time at MediaTek (2021/22) and Logitech (2020).

Research interests: I’m broadly interested in all aspects of machine learning, and particularly in Deep Learning and Human-centered AI (e.g., robustness, fairness, visualization, healthcare, robotics, etc.) I want to develop robust machine-learning applications that can reliably interact with the dynamic world and keep environment and humanity at its core. This goal can be approached from different angles, including ways to make decision-making in AI systems more transparent, methods to quantify uncertainty and improve robustness, innovations to reduce power consumption, and user studies to understand human behavior, trust and acceptance.

latest news

02/2023 My first paper accepted to CVPR 2023 (acceptance rate = 2360/9155 = 25.78%)
12/2022 Accepted ICML 2023 reviewer invitation
11/2022 Presenting “CARBEN: Composite Adversarial Robustness Benchmark” for two sessions at NeurIPS 2022 IBM Research booth
11/2022 One paper accepted to NeurIPS 2022 Workshop on ML for Systems
10/2022 One paper accepted to AAAI 2023 demo track
10/2022 Beginning my visit to IBM Thomas J. Watson Research Center, Yorktown Heights, NY, USA
09/2022 Give an oral presentation for our paper at AutomotiveUI 2022
07/2022 Give an oral presentation/demonstration for our paper at IJCAI 2022
06/2022 Accepted KDD 2022 Workshop on Adversarial Learning Methods for Machine Learning and Data Mining PC invitation
06/2022 Selected as a Honorary Member of The Phi Tau Phi Scholastic Honor Society of R.O.C.
06/2022 Accepted ICML 2022 Workshop on Adversarial Machine Learning Frontiers PC invitation
04/2022 One paper accepted to IJCAI 2022 demo track

selected publications

  1. Towards Compositional Adversarial Robustness: Generalizing Adversarial Training to Composite Semantic Perturbations
    Lei Hsiung, Yun-Yun Tsai, Pin-Yu Chen, and Tsung-Yi Ho
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2023)
  2. A Lab-Based Investigation of Reaction Time and Reading Performance Using Different In-Vehicle Reading Interfaces during Self-Driving
    Lei Hsiung, Yung-Ju Chang, Wei-Ko Li, Tsung-Yi Ho, and Shan-Hung Wu
    In Proceedings of the 14th International ACM Conference on Automotive User Interfaces and Interactive Vehicular Applications, AutomotiveUI (2022)
  3. NeuralFuse: Improving the Accuracy of Access-Limited Neural Network Inference in Low-Voltage Regimes
    Hao-Lun Sun,  Lei Hsiung, Nandhini Chandramoorthy, Pin-Yu Chen, and Tsung-Yi Ho
    Under Review. Preliminary version in NeurIPS workshop. (2022)
  4. NCTV: Neural Clamping Toolkit and Visualization for Neural Network Calibration
    Lei Hsiung, Yung-Chen Tang, Pin-Yu Chen, and Tsung-Yi Ho
    In Proceedings of the 37th AAAI Conference on Artificial Intelligence Demo Track, AAAI (2023)
  5. CARBEN: Composite Adversarial Robustness Benchmark
    Lei Hsiung, Yun-Yun Tsai, Pin-Yu Chen, and Tsung-Yi Ho
    In Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence Demo Track, IJCAI (2022)