Thanks to the chair I hold in Trustworthy AI, I have a number of very interesting and challenging open positions for Masters interships and PhD students.
These positions will be open in january, 2023 for the masters and september, 2023 for the PhDs. If you are interested in Hybrid AI in order to enforce trust on top of black-boxes, please, don’t hesitate to contact me.
Internship 1/3 (February 2023)
Trustworthy AI: Bounding Machine Learning Decisions by Proven Systems (Master + PhD )
In order to achieve the best possible accuracy, decision/recommendation systems built with the help of machine learning act mainly on one lever: the computational complexity of the learned functions (e.g. the number of learned parameters). It is thus common practice to build recommendation systems involving millions of calculations leading to a decision, which has the immediate effect of making it impossible, by construction, to inspect their decision in simple and understandable terms. Aiming solely at precision may have immense applicative interests, but as soon as the implementation requires understanding, justifying, explaining or guaranteeing certain decisions, we see that precision alone is not enough when trust is needed.
In this Master’s internship (followed by a PhD), we propose to study how to bound the behavior of a neural network by two functions, built on some formal language (propositional logic, temporal logic), and adapted to formal verification, in order to mix precision and trust.
The page describing the detailed topic for this internship
The PDF file describing this internship is availaible here
Contact me ASAP if you are interested (contact)
Internship 2/3 (February 2023)
Learning trustworthy systems, a neuro-symbolic approach
In this M2 internship (for which the grant for a PhD is already secured) we propose to study a hybrid approach to machine learning that allows to directly learn functions with structural or semantic properties that would allow to apply, in practice, automatic methods for proof and/or logical reasoning. We will will also study how prior knowledge can be used to converge more quickly or to guarantee that the learned functions respect the properties known in advance.
The page describing the detailed topic for this internship
The PDF file describing this internship is availaible here
Contact me ASAP if you are interested (contact)
Internship 3/3 (February 2023)
Reinforcement learning and Attention Networks for SAT solving
Despite breathtaking results in pattern recognition and signal analysis (sound, image, video, ….) and text, ML / RL-based approaches have not yet succeeded in overtaking conventional methods in solving NP-Hard problems. This class of problems, at the center of theoretical computer science, captures the hardness of a many subfields of AI since its beginnings, such as planning, constraints programming, or even the SAT problem itself (SAT is for “Satisfiability testing”, the main goal of which is precisely to tackle this central problem).
The goal of this Internship (preferably followed by a thesis currently being set up via a CIFRE mechanism) is to study how recent progress allows us to consider the use and extension of ML/RL techniques for the practical resolution of the SAT problem.
The page describing the detailed topic for this internship
Contact me ASAP if you are interested (contact)