Hao-Chun Yang

Senior AI/ML Engineer @ Aptiv

prof_pic.png

Hey, thanks for stopping by! 👋 I’m Hao-Chun Yang, a Senior Machine Learning Engineer at Aptiv, Germany. I lead a team building an end-to-end multimodal sensor-fusion perception system for Level 4 autonomous driving — architecting and optimizing Transformer-based 3D perception with TensorRT and ONNX for edge deployment.

Previously I was a Postdoctoral Researcher at the University of Tübingen / Donders Institute (MHM-Lab, PI: Thomas Wolfers), working on self-supervised and unsupervised representation learning for long-tailed neuroimaging — including a masked mesh vision transformer that improved psychosis detection by 6.7% AUROC and vision-language foundation-model fine-tuning for neurodegenerative disease detection. I received my PhD in Electrical Engineering from National Tsing Hua University, Taiwan supervised by Prof. Chi-Chun Lee.

My interests lie broadly at the intersection of robust and reliable ML: Large-scale Self-Supervised Learning, Generative Models, Domain Adaptation, and Anomaly Detection applied to multi-modality signals.

news

Jul 01, 2024 Excited to join Aptiv as Senior AI/ML Engineer — building the L4 Perception stack (multimodal sensor fusion for self-driving)! 🚗
Mar 01, 2023 After three months of making friends with the immigration office, I have finally embarked on my postdoc journey in Tübingen! Guten Tag! 🍺
Oct 15, 2022 Completed military service. I’m happy to have gained a some weight 🪖
Apr 07, 2022 👨‍🎓 Survived from thesis defense “Physiological-based Affective Computing using Personality Invariant Learning”
Sep 05, 2021 🎉 Best Challenge Poster in 2021 PhysioNet/CinC Challenge (with InventecAI, preprint).

selected publications

  1. ISBI24
    Learning Cortical Anomaly through Masked Encoding for Unsupervised Heterogeneity Mapping
    Hao-Chun Yang, Thomas Wolfers, Ole Andreassen, and 3 more authors
    In 2024 IEEE International Symposium on Biomedical Imaging (ISBI), 2024
  2. A Media-Guided Attentive Graphical Network for Personality Recognition Using Physiology
    Hao-Chun Yang and Chi-Chun Lee
    IEEE Transactions on Affective Computing, 2023
  3. A Mixed-Domain Self-Attention Network for Multilabel Cardiac Irregularity Classification Using Reduced-Lead Electrocardiogram
    Hao-Chun Yang, Wan-Ting Hsieh, and Trista Pei-Chun Chen
    In Computing in Cardiology, CinC 2021, Brno, Czech Republic, September 13-15, 2021
  4. Federated Learning via Conditional Mutual Learning for Alzheimer’s Disease Classification on T1w MRI
    Ya-Lin Huang, Hao-Chun Yang, and Chi-Chun Lee
    In 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society, EMBC 2021, (Virtual), Oct 31 - Nov 4, 2021, 2021