Quentin Garrido, PHD student at ESIEE Paris receives an "outstanding paper honorable mention”

The ICLR is the first international gathering of professionals dedicated to advancing the branch of artificial intelligence known as ‘representation learning’, as well as ‘deep learning’. For the 2023 conference around 5000 papers were submitted, 1500 were accepted and 9 were awarded prizes. 

A particularly remarkable article

His article titled “On the duality between contrastive and non-contrastive self-supervised learning” was praised for its clarity, insight, creativity and the potential to make a lasting impact. 

Quentin Garrido cowrote the article with Yubei Chen, Adrien Bardes, Laurent Najman and Yann LeCun. 

In a nutshell…

The source of the unreasonable effectiveness of self-supervised learning methods is under-researched, questioned, and in particular there are distinct families/lines of methods that appear to have nothing in common but work in a similar way in practice. This article shows that they all have something non-obvious in common.

A quote from a member of the selection committee: “I think that this should be read by anyone working on unsupervised learning.” 

To find out more and download the article

A remarkable journey at ESIEE Paris

Quentin Garrido also won the “Ian Lawson Van Toch Memorial Award for Outstanding Student Paper” at the ISMB 2022 conference for his research in computational biology which he carried out during his 4th year at ESIEE Paris. 

He is currently completing his doctorate at META (Facebook), under the supervision of: 

• Laurent Najman, Professor-Researcher at ESIEE Paris, artificial intelligence specialist.
• Yann LeCun, Scientific Director of Artificial Intelligence at META (Facebook), Professor at University of New York, ESIEE Paris graduate ‘83, Turing Award winner 2019.