I am a master student from KTH Royal Institute of Technology in the program of Master of System, Control and Robotics (my academic email in KTH: yuhanx@kth.se) from 2022 through the 3+2 program between Zhejiang University and KTH, and expect to graduate in July 2024. Currently I am doing my master thesis in EPFL (my academic email in EPFL: yuhan.xie@epfl.ch) , lab LTS4.
In 2023, I obtained my bachelor degree in Automation in Zhejiang University with the honor of “outstanding graduate students”, my bachelor thesis is about long-term forecastiing of the burn through point in the sintering process by deep learning method, supervised by Prof. Xinmin Zhang. During my bachelors, I am also interested in the RoboCup competition. I was a team member of the ZJUNlict from Autumn 2021 to Summer 2022.
In the same year, I was nominated the KTH-One-Year-Scholarship, and went to EPFL from the Autumn 2023 through the SEMP exchange program. I joined LTS4 as a project student, and finished my semester project with William Cappelletti on graph representation learning for epilepsy diagnosis. My master thesis is about improving the alignment between self-supervised learning task and downstream tasks in time series, supervised by Prof. Pascal Frossard (thesis superviser from EPFL) and Prof. Cristian Rojas (thesis examiner from KTH).
Research Interests
Currently, I am interested in interpretable deep learning and machine learning. Specifically, I have research experience in time series forecasting, classification, detection as well as self-supervised learning and transfer learning. I am familiar with Pytorch framework and like to work with encoder-decoder structures, recurrent neural networks, transformers and graph neural networks. I am interested in the application of deep learning models in interdisciplinary areas, like AI for Health, AI for Science, and AI for Industry.
Besides deep learning, I am also interested in multi-agent system. I took courses in applied estimation, robot planning and multi-agent planning, and the most interest topic for me would be designing tools to analyze and help the decision making of multi-agent systems.
Other Backgrounds
you can take a look at my education background, research projects, course projects and other experiences from my CV.
Personal Hobbies
I like travelling, reading and photographing very much. For sports, I like to play tennis and football, and also I am a big football fan!
News
- 2024.08: My work of my bachelor thesis with Bocun EnvFormer: A Decomposition-based Transformer for Multi-step Burn-through Point Prediction in Sintering Process was accepted by IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT!
- 2024.06: I will join EPFL INL lab as an research intern!
- 2024.03: My work of my semester project with William Learning Self-Supervised Dynamic Networks for Seizure Analysis was accepted by the ICLR Workshop of TS4H!
- 2024.02: Passed the doctoral admission committee of EDEE@EPFL!
- 2023.09: Start the master exchange in EPFL
- 2023.06: Graduate from Zhejiang University!
- 2023.05: The paper A Decomposition-based Encoder-Decoder Framework for Multi-step Prediction of Burn-Through Point in Sintering Process was accepted by IEEE ICPS!
- 2022.09: Start the master program in KTH