I am a Lecturer in Computer Science at the University of Exeter, previously I worked at the Chinese Academy of Sciences and the University of Liverpool. I earned my PhD from the University of Liverpool, under the guidance of Prof. Xiaowei Huang and Prof. Xinping Yi.

I am an enthusiastic and dedicated researcher doing research on developing efficient and reliable algorithms for modern machine learning models and applications. I focus on key aspects such as generalisation, robustness, privacy, and interpretability, always striving to provide formal theoretical guarantees and analyses. Additionally, I am actively engaged in the emerging field of building trustworthy Large Language Models (LLMs).

I am pleased to announce multiple openings for PhD and internship positions for the 2024/25 academic year. Scholarships are available through the University of Exeter and the CSC-Exeter program. If you are interested in these opportunities or would like to discuss potential collaborations, please don’t hesitate to contact me at g.jin@exeter.ac.uk.

Funding & Grants

• Trustworthy Intelligent Algorithms in Complex Environments (Co-I) 2022 - 2027
  Funded by the Chinese Academy of Sciences. CN¥20M (≈£2.2M)
• FOCETA (Foundations for Continuous Engineering of Trustworthy Autonomy) (Research Assistant) 2021 - 2023
  Funded by EU H2020. €4,985,540
• EnnCore (End-to-End Conceptual Guarding of Neural Architectures) (Research Assistant) 2020 - 2024
  Funded by EPSRC. £1,721,560
• SOLITUDE (Safety Argument for Learning-enabled Autonomous Underwater Vehicles) (Research Assistant) 2020 - 2022
  Funded by UK DSTL. £400K

Publications

• Out-of-Bounding-Box Trigger: A Stealthy Approach to Cheat Object Detector, ECCV 2024
  Tao Lin, Lijia Yu, Gaojie Jin, Renjue Li, Peng Wu, Lijun Zhang
• Position: Building Guardrails for Large Language Models Requires Systematic Design, ICML 2024
  D Yi, R Mu, G Jin, Y Qi, J Hu, X Zhao, J Meng, W Ruan, X Huang
• TrajPAC: Towards Robustness Verification of Pedestrian Trajectory Prediction Models, ICCV 2023
  L Zhang, N Xu, P Yang, G Jin, CC Huang, L Zhang
• SAFARI: Versatile and Efficient Evaluations for Robustness of Interpretability, ICCV 2023
  Wei Huang, Xingyu Zhao, Gaojie Jin, Xiaowei Huang
• Randomized Adversarial Training via Taylor Expansion, CVPR 2023
  Gaojie Jin, Xinping Yi, Dengyu Wu, Ronghui Mu, Xiaowei Huang
• Certified Policy Smoothing for Cooperative Multi-Agent Reinforcement Learning, AAAI 2023
  Ronghui Mu, Wenjie Ruan, LS Marcolino, Gaojie Jin, Qiang Ni
• Weight Expansion: A New Perspective on Dropout and Generalization, TMLR 2022
  Gaojie Jin, Xinping Yi, Pengfei Yang, Lijun Zhang, Sven Schewe, Xiaowei Huang
• Enhancing Adversarial Training with Second-Order Statistics of Weights, CVPR 2022
  Gaojie Jin, Xinping Yi, Wei Huang, Sven Schewe, Xiaowei Huang
• How does Weight Correlation Affect the Generalisation Ability of Deep Neural Networks, NeurIPS 2020
  Gaojie Jin, Xinping Yi, Liang Zhang, Lijun Zhang, Sven Schewe, Xiaowei Huang

Teaching

• ECM1416: Computational Mathematics

Academic Service

• Reviewer
  ICML, NeurIPS, ICLR, AAAI, CVPR, ICCV, ECCV, JMLR, IJCV, TMLR
• Organise the workshop TrustRL: Trustworthy in Reinforcement Learning at ATC 2024