About Me

Before the more serious part, let me talk about my hobbies : I am a fan of sci-fi litterature and an audiophile. I like swimming and badminton. And I play ukulele.

Education

• 2018 - now : PhD student at Technical University of Munich, Germany
• 2016 - 2018: MSc Informatics at Technical University of Munich, Germany
• 2013 - 2016: Engineering degree at École polytechnique, France
• 2011 - 2013: Preparatory classes at Lycée Louis-le-Grand, France

Work experience

• May 2023 - Jun 2023: Visiting scholar at RIKEN Approximate Bayesian Inference team, Japan
• Feb 2018 - Sep 2018: Working student at Artisense, Germany
• Mar 2017 - Jan 2018: Working student at TUM Computer Vision Group, Germany
• Mar 2016 - Aug 2016: Research internship at Technicolor R&D France, France
• Jun 2015 - Aug 2015: Internship at Valeo Lighting Hubei Technical Center, China

Teaching experience

You can visit my CVG page for a list of my teaching activities at TUM. In summary:

• Teaching assistant for lectures:
  • “Convex Optimization for Machine Learning and Computer Vision” (WS18)
  • “Probabilistic Graphical Models in Computer Vision” (SS19)
• Co-organize practical courses and seminars:
  • “Learning For Self-Driving Cars and Intelligent Systems” (practical course, WS19)
  • “Beyond Deep Learning: Uncertainty Aware Models” (practical course, SS20)
  • “Beyond Deep Learning: Selected Topics on Novel Challenges” (seminar WS20-SS22)

Additionally, I have also supervised several individual student projects including master thesis, interdisciplinary project and guided research, some of which results in publications.

Publications

You can visit my Google Scholar for a list of my publications.

• Cong B, Daheim N, Shen Y, Cremers D, Yokota R, Khan ME, Möllenhoff T. “Variational Low-Rank Adaptation Using IVON”. In NeurIPS Workshop on Fine-Tuning in Modern Machine Learning: Principles and Scalability (FITML). 2024
• Shen Y*, Daheim N*, Cong B, Nickl P, Marconi GM, Bazan C, Yokota R, Gurevych I, Cremers D, Khan ME, Möllenhoff T. “Variational Learning is Effective for Large Deep Networks”. In International Conference on Machine Learning (ICML). 2024 (spotlight)
• Koke C, Saroha A, Shen Y, Eisenberger M, Cremers D. “ResolvNet: A Graph Convolutional Network with multi-scale Consistency”. In NeurIPS Workshop: New Frontiers in Graph Learning. 2023
• Tomani C*, Waseda FK*, Shen Y, Cremers D. “Beyond In-Domain Scenarios: Robust Density-Aware Calibration”. In International Conference on Machine Learning (ICML). 2023
• Hsu HH*, Shen Y*, Tomani C, Cremers D. “What Makes Graph Neural Networks Miscalibrated?”. Advances in Neural Information Processing Systems (NeurIPS). 2022
• Shen Y, Cremers D. “Deep Combinatorial Aggregation”. Advances in Neural Information Processing Systems (NeurIPS). 2022
• Hsu HH*, Shen Y*, Cremers D. “A Graph Is More Than Its Nodes: Towards Structured Uncertainty-Aware Learning on Graphs”. In NeurIPS Workshop: New Frontiers in Graph Learning. 2022
• Wang Y, Shen Y, Cremers D. “Explicit Pairwise Factorized Graph Neural Network for Semi-Supervised Node Classification”. In Uncertainty in Artificial Intelligence (UAI). 2021
• Yan J, Peng Z, Yin H, Wang J, Wang X, Shen Y, Stechele W, Cremers D. “Trajectory prediction for intelligent vehicles using spatial‐attention mechanism”. IET Intelligent Transport Systems. 2020
• Shen Y, Demarty CH, Duong NQ. Deep learning for multimodal-based video interestingness prediction. In 2017 IEEE International Conference on Multimedia and Expo (ICME). 2017

(* denotes equal contribution)

Skills

Languages

Chinese (mother tongue), English (fluent), French (fluent), German (intermediate), Japanese (basic notions)

Programming languages

• Currently use a lot: Python
• Have done projects with: Java, C/C++, CUDA, OCaml, Visual Basic, Pascal, HTML/CSS/Javascript

Softwares & frameworks

PyTorch (contributor), Jax, LaTeX, OpenCV, Jekyll, Git/SVN, CATIA, image/audio/video editing, soundtrack mixing, Linux, office suite