Illustration transfert



Axis 3: Transfer learning and generalization

The goals of this research axis are to transfer and exploit the results and knowledge acquired to other levels of scale, organisms or systems that have been less studied or only partially observed, as well as to develop more robust comparative approaches that better exploit the generic scope of the data.

In direct relation with the research supported in axes 1 and 2, the goals are as follows :

  • Adapt integrative and predictive approaches to situations of partial and incomplete information on the target mechanisms or processes, both to test the generic nature of their conclusions and extend their applications;
  • Validate predictions made in model systems by comparing them with other organisms;
  • Compare and exploit different biological systems to broaden the scope;
  • Develop the digital methods needed to extrapolate to the whole organism from observations made in vitro (cells/organs), a major challenge in the context of a reduction in in vivo studies in animal husbandry and toxicology.

The modelling and learning approaches mentioned in axes 1 and 2 can be mobilised, but with specificities linked to transfer, which requires the reliability of predictions to be adapted and evaluated.

In this folder

Photo vache dans une étable
In plant and animal genetics, selection programmes aim to identify individuals whose performance (yield, resistance to disease or environmental stress) meets previously defined criteria. This selection requires the acquisition of data, in the field or in breeding, which can be costly or time-consuming.

Modification date: 06 October 2023 | Publication date: 22 March 2022 | By: Com