About us:

The Mat-Tech Lab (Mathematical Technologies Lab) is a vibrant interdisciplinary team composed by mathematicians, engineers, physicists and neuroscientists working at the interface between the development of new technologies in healthcare (& beyond) and mathematics. The goal of the lab is to promote new innovations in the field of healthcare technologies, while being able to re-adapt creatively the modelling frameworks used up to now. We value industrial partnership and hybrid developments (across academia and industry).  

News and events

Machine Unlearning: Definitions, Methods, and Auditing for Trustworthy AI. 

Abstract. Machine learning models revolutionize how we interact with technology. However, as models are trained on increasingly large and potentially sensitive datasets, they become subject to privacy concerns, biases, and the retention of outdated information. Machine unlearning emerges as a critical field to address these issues by selectively removing training data's influence from a trained model. This talk will begin by outlining the crucial motivations driving the need for machine unlearning. We'll explore scenarios where unlearning protects user privacy, corrects training data errors, and ensures ongoing model relevance. Next, we'll provide mathematically sound definitions of unlearning, and explore the similarities and differences with the field of differential privacy. Next, the focus shifts to existing approaches to machine unlearning. We'll examine techniques like influence functions, retraining strategies, and knowledge distillation.  Crucially, the talk will address the importance of auditing unlearning algorithms. We'll discuss metrics for evaluating unlearning effectiveness,  guarantees of privacy protection, and the potential for unintended consequences.

Thanks to:

School of Engineering,  Institute of Systems Engineering,  HES-SO Valais-Wallis – We thank School of Engineering of the HES-SO Valais-Wallis for welcoming us and support our research projects. 

We thank the Sense Innovation and Research Center for the highly interdisciplinary environment it creates for their members.