Scholar
Hanjiang Hu
Google Scholar ID: UyooQDYAAAAJ
Carnegie Mellon University
Citations & Impact
Citations
661
H-index
14
i10-index
16
Publications
20
Co-authors
41
list available
Publications
Resume (English only)
Academic Achievements
- 1. Selected as 2025 DAAD AINeT fellow on Explainable AI; 2. Gave an invited talk at 2025 INFORMS Annual Meeting; 3. One paper about verifiable safety Q-filter accepted to L-CSS; 4. Selected as 2025 ASME Dynamic Systems & Control Division (DSCD) Rising Star; 5. Organizing Workshop on Foundation Models for Control (FM4Control) at MECC 2025; 6. Gave a talk on 'Verified Dynamical Safety with Neural Barrier Functions: From Dynamical Systems to Language Models' at Northeast Systems and Control Workshop 2025; 7. One paper about verified neural HJ-reachability value function accepted to JAIR; 8. Our neural network verification toolbox ModelVerification.jl accepted to CAV 2025; 9. Organizing the RoboSense Challenge at IROS 2025; 10. One paper about safe neural PDE control accepted to L4DC 2025; 11. One paper about sensor-placement-aware robust perception accepted to NeurIPS 2024 as a Spotlight; 12. One paper about verification of neural CBF accepted to CoRL 2024; 13. Released our new comprehensive NN verification toolbox ModelVerification.jl; 14. One paper about safe reinforcement learning accepted to DMLR; 15. One paper about safe control and NN verification accepted to L4DC 2024; 16. Gave a talk about safe control and verification of neural dynamics in CMU Learning and Control Seminar; 17. One paper about sensor-placement-aware robust perception accepted to ICRA 2024; 18. One paper about robustness certification accepted to AISTATS 2024; 19. Organizing the RoboDrive Challenge at ICRA 2024; 20. One paper about out-of-distribution robust perception accepted to NeurIPS 2023; 21. One paper about cross-domain robust perception accepted to IROS 2023; 22. One paper about safe reinforcement learning accepted to ICML 2023; 23. Received a fellowship from the Machine Learning Department, School of Computer Science at CMU; 24. Organizing the RoboDepth (Robust Out-of-distribution Depth prediction) Challenge at ICRA 2023; 25. One paper about robustness certification accepted to CoRL 2022; 26. One paper about safety benchmark in autonomous driving accepted to NeurIPS 2022.
Research Experience
- 1. Conducting doctoral research at the Robotics Institute, CMU, focusing on provable safety and robustness guarantees for learning-enabled autonomous systems; 2. Involved in multiple research projects, including autonomous driving, robotics, PDE-governed complex systems, and multi-turn interactions with large language models (LLMs).
Education
- 1. PhD candidate in the Department of Electrical and Computer Engineering (ECE) at CMU, advised by Prof. Changliu Liu; 2. Master's degree in Machine Learning (MSML) at CMU; 3. B.Eng. and M.S. from Shanghai Jiao Tong University (SJTU), advised by Prof. Hesheng Wang.
Background
- Research interests: building provable safety and robustness guarantees for learning-enabled autonomous systems, particularly at the intersection of robotics, control theory, and machine learning. Specifically, developing formal verification methods to certify robustness against semantic perturbations in perception models and verifying neural control barrier functions to ensure forward invariant safety in learning-based control and dynamics.
Miscellany
- On the job market and open to opportunities. Please feel free to reach out if my research aligns with your interests.
