About Me
I am Cyril Shih-Huan Hsu, a research scientist working at the intersection of evolutionary computation, deep reinforcement learning, generative models, and multi-agent intelligent systems design. I completed my PhD in computer science at the University of Amsterdam, supervised by Dr. Paola Grosso and Dr. Chrysa Papagianni.
My research focuses on developing and applying state-of-the-art ML algorithms to address complex, real-world problems. I am particularly interested in how adaptation, cooperation, and emergence can guide the design of intelligent, data-driven systems.
My background spans both academia and industry, including research in scalable optimization, applied AI development, and international startup experience arcoss Asia, Europe, and Central America. I am currently a postdoctoral researcher at the University of Amsterdam, where I explore LLM-based agentic AI and context engineering for decentralized multi-agent systems in next-generation networking.
When not debugging code or trapping myself in difficult math derivations,
I try to live by a simple belief:
If you work really hard and you're kind, amazing things will happen.
— Conan O’Brien.
Applied Machine Learning in Industry
As a Machine Learning Researcher & Engineer at Emotibot Technologies, I worked on production-facing intelligent systems and embedded deep learning.
- Real-time facial recognition systems
- Multimodal affective computing
- Deep neural networks for embedded systems
Building an AI Healthcare Startup
As CTO & Co-founder of Deep Cube SA, I led the development of AI systems for healthcare, including glaucoma screening, skin-disorder analysis, and clinical data collection.
This period overlapped with my role as Technical Director of AI at Relajet Technologies in Taipei (2018–2020).
Research in Intelligent Networked Systems
At the University of Amsterdam, I completed a PhD in Informatics and now work as a postdoctoral researcher on agentic AI, context engineering, and decentralized multi-agent systems.
- PhD researcher, teaching, and student thesis supervision (2021–2025)
- Visiting researcher at NEC Laboratories Europe in Heidelberg (2024)
- Postdoctoral researcher in agentic AI-driven network automation (2025–present)
Selected Publications
Journal Papers
- Hsu, C. S. H., Dalgkitsis, A., Grosso, P., & Papagianni, C. (2026). Transformer-Empowered Actor-Critic Reinforcement Learning for Sequence-Aware Service Function Chain Partitioning. IEEE Transactions on Network Science and Engineering (TNSE), 2026. [Link]
- Hsu, C. S. H., Martín-Pérez, J., De Vleeschauwer, D., Valcarenghi, L., Li, X., & Papagianni, C. (2025). A Deep RL Approach on Task Placement and Scaling of Edge Resources for Cellular Vehicle-to-Network Service Provisioning. IEEE Transactions on Network and Service Management (TNSM), 2025. [Link]
- Hsu, C. S. H., De Vleeschauwer, D., & Papagianni, C. (2023). SLA Decomposition for Network Slicing: A Deep Neural Network Approach. IEEE Networking Letters (NL), 2023. [Link]
Conference Papers
- Hsu, C. S. H., Cheng, W. Y. C., & Papagianni, C. (2026). Progressive Semantic Communication for Efficient Edge-Cloud Vision-Language Models. 2026 IEEE Global Communications Conference (GLOBECOM). [Link]
- Hsu, C. S. H., Li, X., Zanzi, L., Yang, Z., Papagianni, C., & Costa-Pérez, X. (2026). MapViT: A Two-Stage Vision Transformer-Based Framework for Real-Time Radio Quality Map Prediction in Dynamic Environments. 2026 IEEE International Conference on Communications (ICC). [Link]
- Dalgkitsis, A., Hsu, C. S. H., Papagianni, C., Grosso, P. (2026). LLM-ASF6G: LLM-Based Algorithm Selection Framework for 6G Network Service Optimization. 2026 Joint European Conference on Networks and Communications & 6G Summit (EuCNC). [Link]
- Hsu, C. S. H., Papagianni, C., & Grosso, P. (2025). RAILS: Risk-Aware Iterated Local Search for Joint SLA Decomposition and Service Provider Management in Multi-Domain Networks. 2025 IEEE Conference on High Performance Switching and Routing (HPSR). [Link]
- Dalgkitsis, A., Hsu, C. S. H., Papagianni, C., Grosso, P., & de Laat, C. (2025). LLM-based Optimization Algorithm Selection for High-Performance Networks Orchestration. Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC Workshops). [Link]
- Hsu, C. S. H., De Vleeschauwer, D., Papagianni, C., & Grosso, P. (2025). Online SLA Decomposition: Enabling Real-Time Adaptation to Evolving Network Systems. 2025 Joint European Conference on Networks and Communications & 6G Summit (EuCNC). [Link]
- Hsu, C. S. H., Martín-Pérez, J., Papagianni, C., & Grosso, P. (2023). V2N Service Scaling with Deep Reinforcement Learning. 2023 IEEE Conference on Network Operations and Management Symposium (NOMS). [Link]
- Sun, M. C., Hsu, C. S. H., Yang, M. C., & Chien, J. H. (2018). Context-Aware Cascade Attention-Based RNN for Video Emotion Recognition. 2018 First Asian Conference on Affective Computing and Intelligent Interaction (ACII Asia). [Link]
- Chang, W. Y., Hsu, C. S. H., & Chien, J. H. (2017). FATAUVA-Net: An Integrated Deep Learning Framework for Facial Attribute Recognition, Action Unit Detection, and Valence-Arousal Estimation. 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). [Link]
- Hsu, C. S. H., & Yu, T. L. (2015). Optimization by Pairwise Linkage Detection, Incremental Linkage Set, and Restricted/Back Mixing: DSMGA-II. Proceedings of the 2015 Annual Conference on Genetic and Evolutionary Computation (GECCO). [Link]
Contact
s.h.hsu[AT]uva.nl · Google Scholar · LinkedIn