Escola Paulista de Medicina
FarmoLAB

Webinar - Data Driven Federated Learning: A new privacy preserving method for sharing knowledge

On April 26, at 1 pmThierry Hansen

Moderator: Mirian Hayashi, PhD, Associate Professor, Department of Pharmacology, Escola Paulista de Medicina - Unifesp

Artificial Intelligence (AI) has become a powerful research catalyst in science. At the core of modern AI is the ability to automatically extract knowledge from data and build accurate predictive models. To maximize this impact, it is critical to have access to enough good quality data to allow machine learning algorithms to extract relevant knowledge and produce useful models. One of the main challenges in AI is therefore to compile such pivotal datasets, which is particularly difficult in drug discovery due to the confidential nature of the primary information: the chemical structure. Even with the availability of public data, the most valuable knowledge remains embedded and locked in private silos despite the willingness of industry to share non-competitive information. This presentation describes an approach to facilitate knowledge sharing between organisation, whilst maintaining the confidentiality of private data containing this knowledge.

Speaker: Thierry Hanser - Head of Molecular Informatics and AI group - Lhasa Limited, UK

Little chronology :

My background is in Cheminformatics, I did my PhD in Strasbourg in 1994 on Machine Learning: “Automatic extraction of reaction knowledge from reaction databases”.In 1995, I did a first Post-Doc at Harvard university (Chemistry Department) developing retro-synthesis in-silico methods in Prof. E.J. Cory’s group.
In 1997, I did a second Post-Doc at Leeds university (Chemistry Department) contributing to De Novo Ligand Design in Prof. Peter Johnson’s groupLater I created my own company (IXELIS) developing Pharmaceutical information systemsIn 2006 I joined Lhasa Limited in Leeds where I initially contributed in the design and development of Lhasa’s Cheminformatics platform. My research activity includes bridging Cheminformatics, Machine Learning and AI in order to design new Knowledge Discovery and predictive modelling methodologies.After 16 years at Lhasa, I am leading the Molecular Informatics and AI group.

Registration to the event: link

Link to watch the event: unifesp.br/webinar

Speaker further information can be found at: link

© 2013 a 2024 - Universidade Federal de São Paulo - Unifesp
Campus São Paulo - Rua Botucatu, n° 740 - 5º andar - Vila Clementino - São Paulo - SP - CEP: 04023-062 

Please publish modules in offcanvas position.