Dear all,
Apologies for cross-posting.
Subject:
All the positions will be a part of the ERC Starting Grant (2021) - DYNASTY (Dynamics-Aware Theory of Deep Learning) project, which focuses on various aspects of Deep Learning Theory. The potential topics include (and are not limited to):
- Generalization bounds for neural networks
- Implicit regularization
- Non-convex optimization
- Network compression
Candidate Profile:
- For the Internship positions: Enrollment at a masters/PhD program is required, preferably pure/applied mathematics or statistics.
- For the Phd positions: A masters degree in applied mathematics, statistics, or computer science is required. (A direct PhD after undergrad is not possible).
- For the Postdoc positions: PhD degree (and publications) in ML Theory or related fields such as applied mathematics (preferably applied probability, stochastic analysis, operations research), statistics, computer science.
A background on dynamical systems / fractal geometry / heavy-tailed stochastic recursions will be a plus.
I will not consider candidates that only have an applications background.
The SIERRA Team:The researcher will be hosted at the
INRIA Paris center, in downtown Paris.
A non-exhaustive list of topics that are currently investigated by the
researchers of the SIERRA team includes:
- Large scale non-smooth optimization methods.
- Distributed optimization methods (asynchronous stochastic gradient optimization).
- First order-methods.
- Nonconvex optimization.
- Kernel methods.
Application Process / Deadline:
- There is no specific deadline. The applications will be considered on a "first-come-first-serve" basis.
- Internships typically take 3-6 months. The timing can be flexible.
- In the application, please send a
resume, a
motivation letter, and at least one
reference letter to
Umut Simsekli -- "umut.simsekli at
inria.fr".
Best regards,
Umut Simsekli
--