Engineer F/M for Neural Architecture Growth

Created in 2008, the Inria Saclay Center is located at the heart of the Paris-Saclay scientific and technological excellence cluster, which alone accounts for 15% of French research. Serving the development of the Université Paris-Saclay and the Institut Polytechnique de Paris, the Inria Saclay center employs 80 people in research support services and 500 scientists of 54 nationalities.

Benefiting from continuous growth, the center now has a total of 42 project-teams and two in the process of being created, including 21 jointly with the Institut Polytechnique de Paris and 16 with the Université Paris-Saclay.

Contexte et atouts du poste

We are a small team of PhD students, post-docs and permanent members interested into Frugal AI, with expertise in machine learning, optimization, open-source library managing, etc.

We have a GPU cluster, and the national cluster Jean Zay is available as well (with ~3000 GPU cards).

This project is part of a European project, named MANOLO.

Mission confiée

The person recruited will work on Frugal AI, and more precisely on Neural Architecture Growth, both on theoretical and practical aspects, to be able to develop architectures that develop on the fly while training and thus spare computational resources needed by classical Neural Architecture Search (whether manual or automated).

More precisely, the topic will be to develop, implement and test reinforcement learning approaches for neural architecture growth. The person recruited is expected to follow the literature on this topic and suggest development directions.

Principales activités

The main task is to contribute to theoretical developments of the topic (optimization, searching for growth strategies), to the open-source library of the team, and to the team's social and daily life.

It is also expected to take part to writing papers and deliverables that are required by the European project, which correspond to the advancement of the state of the project.

Compétences

Expected skills :

• machine learning (neural networks...)
• reinforcement learning (policy gradient using neural networks, ...)
• maths (statistics, analysis, etc.)
• optimization (gradient descent, ...)
• programming (python, PyTorch)

Avantages

• Subsidized meals
• Partial reimbursement of public transport costs
• Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
• Possibility of teleworking and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities
• Access to vocational training
• Social security coverage

Motivated to work on Frugal AI with a small team including PhD students, post-docs, and permanent researchers.