Sannara EK

Sannara Ek - ឯក សាងណារ៉ា

I am a PostDoctoral Research Associate at the University of Cambridge, working with Prof. Robert Mullins on developing efficient edge-based vision–language models (VLMs) for applications in the Earth Observation domain. I obtained my PhD from Université Grenoble Alpes, where I was supervised by Prof. Philippe Lalanda and Prof. François Portet. My doctoral research focused on Federated Learning in Pervasive Computing environments, developing lightweight machine learning solutions resilient to data hetereogenity. I also hold a Master’s degree in Computer Vision and Robotics from Université Grenoble Alpes.

My research lies at the intersection of mobile and pervasive systems and the design of efficient, robust AI models.

Activities

July 2025 — I have joined the University of Cambridge - Department of Computer Science and Technology where I will be working on efficient VLM edge-systems applied to multi-spectral satellite images.
November 2024 — I have succesfully defended my PhD thesis titled Personalized Federated Learning for Sensor-based Human Activity Recognition in Pervasive Heterogeneous Environments in front of a jury composed of Professors Jiannong Cao, Cecilia Mascolo, Giovanni Neglia, Keiichi Yasumoto, Vania Marangozova and Jean-Michel Tran.
August 2024 — Participated in a research exchange at the Internet and Mobile Computing Laboratory (IMCL) - Hong Kong Polytechnic University to collaborate and extend my FL findings into the vision domain.
October 2022 — Participated in a research exchange at EveryWare Lab - University of Milan to collaborate on findings regarding pre-trained and self-supervised learning models in the wearable Human Activity Recognition domain.
November 2021 — Visited the Measurement and Electronic Architectures group - Institut Jean Lamour to investigate deep learning solutions for arc fault detection.

Journals

	Comparing Self-Supervised Learning Techniques for Wearable Human Activity Recognition Sannara Ek, Riccardo Presotto, Gabriele Civitarese, François Portet, Philippe Lalanda, Claudio Bettini CCF Transactions on Pervasive Computing and Interaction pdf \| code
	Transformer-based models to deal with heterogeneous environments in Human Activity Recognition Sannara Ek, François Portet, Philippe Lalanda Personal and Ubiquitous Computing pdf \| code
	Evaluation and comparison of federated learning algorithms for Human Activity Recognition on smartphones Sannara Ek, François Portet, Philippe Lalanda, German Vega Pervasive and Mobile Computing pdf \| code

Proceedings

	Combining Public Human Activity Recognition Datasets TO Mitigate Labeled Data Scarcity Riccardo Presotto, Sannara Ek, Gabriele Civitarese, François Portet, Philippe Lalanda, Claudio Bettini 2023 IEEE International Conference on Smart Computing (SMARTCOMP) pdf \| code
	Federated Learning Within Pervasive Heterogeneous Environments Sannara Ek, François Portet, Philippe Lalanda 2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops) pdf
	Federated Self-Supervised Learning in Heterogeneous Settings: Limits of a Baseline Approach on HAR Sannara Ek, Romain Rombourg, François Portet, Philippe Lalanda 2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops) pdf
	A Federated Learning Aggregation Algorithm for Pervasive Computing: Evaluation and Comparison Sannara Ek, François Portet, Philippe Lalanda, German Vega 2021 IEEE International Conference on Pervasive Computing and Communications (PerCom) pdf \| code
	Evaluating Federated Learning for Human Activity Recognition Sannara Ek, François Portet, Philippe Lalanda, German Vega 2020 Workshop AI for Internet of Things, in conjunction with IJCAI-PRICAI pdf
	Evaluation of federated learning aggregation algorithms: application to human activity recognition Sannara Ek, François Portet, Philippe Lalanda, German Vega UbiComp/ISWC '20 Adjunct: Adjunct Proceedings of the 2020 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2020 ACM International Symposium on Wearable Computers pdf