thesis on the topic of "Connecting discrete and continuous neural
network models".
The goal of this Master thesis is to establish formal relationships
between the behavior of a discrete neural network and the corresponding
continuous neural ordinary differential equation (neural ODE). Such
relationships could then be used to guarantee the safe behavior of one
of the models after proceeding to the safety verification of the other
model.
The main objectives of this Master thesis are the following:
- Short literature review of existing neural ODE models and their
relation to discrete neural networks, and on formal behavioral
relationships between models.
- Establishing formal relations between the discrete and continuous
neural models.
- Combining these formal relationships with existing verification
algorithms (for neural networks or neural ODE) to deduce the safety of
one model based on the safety verification of the other.
More details on desired profile and application procedure are provided
in the following link:
[ https://gdr-macs.fr/sites/default/files/2024-11/241112_Master_thesis_neural_ODE.pdf | https://gdr-macs.fr/sites/default/files/2024-11/241112_Master_thesis_neural_ODE.pdf ]
Best regards,
_ _ _ _
Abdelrahman Sayed Ibrahim
MSCA-CLEAR-Doc
COSYS Department - ESTAS Laboratory
Campus de Lille
20, rue Élisée Reclus • BP 70317
F-59666 Villeneuve d'Ascq Cedex • FRANCE
**********************************************************
*
* Contributions to be spread via DMANET are submitted to
*
* DMANET@zpr.uni-koeln.de
*
* Replies to a message carried on DMANET should NOT be
* addressed to DMANET but to the original sender. The
* original sender, however, is invited to prepare an
* update of the replies received and to communicate it
* via DMANET.
*
* DISCRETE MATHEMATICS AND ALGORITHMS NETWORK (DMANET)
* http://www.zaik.uni-koeln.de/AFS/publications/dmanet/
*
**********************************************************
Posts
