Education
- 09/2016 - 07/2020, Beijing Institute of Technology, China, B.Eng. in Electronic Information Engineering.
- 11/2020 - 10/2023 (Expected), Technical University of Munich, Germany, M.Sc. in Electrical and Computer Engineering.
Research Experience
- 03/2019 - 06/2020, Bachelor’s thesis, Computer Vision Lab, BIT, Beijing, China
- Developed a defect detection system of steel plates in C++
- Implemented the matching algorithm by extracting the contours of images and comparing the Hausdorff distance between them
- 04/2021 – 10/2021, Research Intern, fortiss GmbH, Munich, Germany
- Implemented a sampling-free Laplace Approximation for Bayesian Neural Network GitHub.
- Studied the effectiveness and computational necessity of a diagonal Hessian approximation in the Laplace approximation on object detection tasks in an autonomous driving scenario.
- Proposed hybrid ViT models with scattering transform to improve performance for ViT on small datasets or frequency-sensitive tasks.
- 04/2022 – 12/2022, Working Student, Machine Learning Research Lab, Volkswagen, Munich, Germany
- Built a Gym interface for the Quanser QCar environment.
- Integrated model-based reinforcement learning and spatial world model to optimize spatial navigation within the QCar environment.
- 01/2023 - now, Master’s thesis, Machine Learning Research Lab, Volkswagen, Munich, Germany
- Worked in information-theoretic spatial exploration within a probabilistic world model, focusing on controlling a mobile agent (e.g., a robot car) to navigate toward unexplored regions in an information-theoretic manner.
- Expanded the exploration framework to accommodate dense 3D environments, emphasizing real-time performance for efficient navigation.
Contact
Email: ziqingzhao.23@gmail.com