Section: Bilateral Contracts and Grants with Industry

Bilateral Grants with Industry

CIFRE contract with Technicolor R&I France on Very large scale visual comparison

Participants : Rémi Gribonval, Himalaya Jain.

• Duration: 3 years (2015-2018)

• Research axis: 3.1.2

• Partners: Technicolor R&I France; Inria-Rennes

• Funding: Technicolor R&I France; ANRT

The grand goal of this thesis was to design, analyze and test new tools to allow large-scale comparison of high-dimensional visual signatures. Leveraging state of the art visual descriptors, the objective is to obtain new compact codes for visual representations, exploiting sparsity and learning, so that they can be stored and compared in an efficient, yet meaningful, way.

CIFRE contract with Facebook Artificial Intelligence Research, Paris on Deep neural networks for large scale learning

Participants : Rémi Gribonval, Pierre Stock.

• Duration: 3 years (2018-2021)

• Research axis: 3.1.2

• Partners: Facebook Artificial Intelligence Research, Paris; Inria-Rennes

• Funding: Facebook Artificial Intelligence Research, Paris; ANRT

The overall objective of this thesis is to design, analyze and test large scale machine learning algorithms with applications to computer vision and natural language processing. A major challenge is to design compression techniques able to replace complex and deep neural networks with much more compact ones while preserving the capacity of the initial network to achieve the targeted task. An avenue primarily envisioned to achieve this goal is to rely on structured linear layers.