C. Castera, J. Bolte, C. Févotte, E. Pauwels
"An Inertial Newton Algorithm for Deep Learning"
Submitted at Journal of Machine Learning Research 2019.
Welcome
I am a second-year Ph.D student within the SC team of IRIT, the computer science research institute of Toulouse. I work under the supervision of Cédric Févotte, Édouard Pauwels and Jérôme Bolte. I am a member of the FACTORY project.
I was previously a six-month intern at ITAV under the supervision of Pierre Weiss and Frédéric de Gournay where I was working on Semi-infinite Programming and Inverse Problems. I both graduated from INSA Toulouse and Université Paul Sabatier where I obtained master's degrees respectively in "Applied Mathematics and Modeling" and "Applied Mathematics, major: Partial Differential Equations".
My research focuses on optimization algorithms and their applications in Deep Learning and Machine Learning models. I also spend a reduced amount of time giving tutorials and practicals at INP ENSEEIHT where I teach Probabilities, Statistics and Integration Theory to master students.
Research Interests
- Deep Learning
- Nonconvex Optimization
- Statistics and Machine Learning Theory
Recent Work
Pytorch Keras and Tensorflow codes for INDIAN algorithm
INDIAN algorithm in a nutshell
Poster presented at NeurIPS 2019
Detailed List of Publications
Contact
E-mail adress:
[firstname].[lastname]@irit.fr
Physical address:
INP-ENSEEIHT, Office F418
2 rue Charles Camichel
Toulouse 31000, France