Self-Supervised learning for Neural Architecture Search (NAS)
This is an incremental approach to improving neural architecture search for AI researchers, as it focuses on using unlabeled data without demonstrating new breakthroughs.
The paper tackled the problem of automating neural architecture search using self-supervised learning with unlabeled data, but it did not report any concrete results or numbers.
The objective of this internship is to propose an innovative method that uses unlabelled data, i.e. data that will allow the AI to automatically learn to predict the correct outcome. To reach this stage, the steps to be followed can be defined as follows: (1) consult the state of the art and position ourself against it, (2) come up with ideas for development paths, (3) implement these ideas, (4) and finally test them to position ourself against the state of the art, and then start the sequence again. During my internship, this sequence was done several times and therefore gives the tracks explored during the internship.