Scholar

Ke Jiang

Google Scholar ID: lJ3AaHoAAAAJ

Nanjing University of Aeronautics and Astronautics / Osaka University

Reinforcement LearningComputer VisionMachine LearningDynamical Systems

Citations & Impact

All-time

Citations

H-index

i10-index

Publications

Co-authors

list available

Contact

Publications

9 items

Browse publications on Google Scholar (top-right) ↗

Resume (English only)

Academic Achievements

- Publications:
- 'Variational OOD State Correction for Offline Reinforcement Learning' accepted by AAAI 2026
- 'Beyond Non-Expert Demonstrations: Outcome-Driven Action Constraint for Offline Reinforcement Learning' accepted by Pattern Recognition
- 'Towards Reliable Offline Reinforcement Learning via Lyapunov Uncertainty Control' accepted by IEEE Transactions on Neural Networks and Learning Systems
- 'RoGA: Towards Generalizable Deepfake Detection through Robust Gradient Alignment' accepted by ICME 2025 (Oral)
- 'Recovering from out-of-sample states via inverse dynamics in offline reinforcement learning' accepted by NeurIPS 2023
- Project Experience: Involved in research on offline reinforcement learning methods and theories in complex real-world scenarios, funded by the National Natural Science Foundation of China, led by Prof. Xiaoyang Tan

Research Experience

- From 14 April, 2025: Working as a visiting researcher at Machine Learning & Systems Laboratory, Graduate School of Information Science and Technology, Osaka University, under the supervision of Professor Yoshinobu Kawahara

Education

- B.Sc.: 2015.9-2019.6, School of Computer Science, Nanjing University of Information Science and Technology
- M.Sc.: 2019.9-2022.4, College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Advisor: Prof. Xiaoyang Tan
- Ph.D. Student: 2022.4-Present, College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Advisor: Prof. Xiaoyang Tan

Background

- Research Interests: (Robust, Generalizable, Safe, Offline) Reinforcement Learning; Generative Models for Long-horizon Planning; Cross-domain Classification (Videos & Image)

Co-authors

2 total