6 papers
Efficient and Adaptive Human Activity Recognition via LLM Backbones
Aleksandr Bredikhin, Philippe Lalanda, German Vega
Human Activity Recognition (HAR) is a core task in pervasive computing systems, where models must operate under strict computational constraints while remaining robust to heterogen…
Sample-Efficient Adaptation of Drug-Response Models to Patient Tumors under Strong Biological Domain Shift
Camille Jimenez Cortes, Philippe Lalanda, German Vega
Predicting drug response in patients from preclinical data remains a major challenge in precision oncology due to the substantial biological gap between in vitro cell lines and pat…
Parameter-Efficient Fine-Tuning for HAR: Integrating LoRA and QLoRA into Transformer Models
Irina Seregina, Philippe Lalanda, German Vega
Human Activity Recognition is a foundational task in pervasive computing. While recent advances in self-supervised learning and transformer-based architectures have significantly i…
Quantifying Uncertainty in Machine Learning-Based Pervasive Systems: Application to Human Activity Recognition
Vladimir Balditsyn, Philippe Lalanda, German Vega +1
The recent convergence of pervasive computing and machine learning has given rise to numerous services, impacting almost all areas of economic and social activity. However, the use…
TaskVAE: Task-Specific Variational Autoencoders for Exemplar Generation in Continual Learning for Human Activity Recognition
Bonpagna Kann, Sandra Castellanos-Paez, Romain Rombourg +1
As machine learning based systems become more integrated into daily life, they unlock new opportunities but face the challenge of adapting to dynamic data environments. Various for…
Microservice-based edge platform for AI services
Philippe Lalanda, German Vega, Denis Morand
Pervasive computing promotes the integration of smart electronic devices in our living and working spaces to provide advanced services. Recently, two major evolutions are changing…