Séminaire de Jonathan KOBOLD (AiDrones GMBH, docteur de l’Université d’Évry Paris-Saclay, sous la direction de Vincent VIGNERON et de Hichem MAAREF), le 5 mars 2020 à 14h, site Pelvoux: «The Receptive Field in CNNs»

Jonathan Kobold a soutenu sa thèse à l’Université d’Évry Paris-Saclay en 2019, sous la direction de Vincent Vigneron et de Hichem Maaref, de l’équipe SIAM de IBISC.

Titre: The Receptive Field in CNNs

Abstract:

Convolutional neural networks (CNNs) are currently the number one tool for pattern recognition, object detection and segmentation. At the core of these network architectures is the concept of the receptive field. If it is too small a CNN architecture will not be able learn its task. Developers of new CNN architectures need a good understanding of the receptive field. It is also useful to determine whether a CNN architecture is able to learn a given task or not. This talk will give an introduction to the receptive field and why it is so important. The different methods for increasing the receptive field are discussed and a method for investigating the receptive field, the effective receptive field, will be presented.

Présentation de Jonathan Kobold (au format PDF)

Short CV:

Jonathan Kobold is an expert for computer vision and machine learning. He originally studied physics with a specialisation on machine learning in Regensburg, Germany. In 2016 he came to France and did his Ph.D. under the supervision of Prof. H. Maaref and V. Vigneron at the laboratoire IBISC. His thesis was in close collaboration with the neurology group of the Centre Hospitalier Sud Francilien (CHSF) under the lead of Prof. D. Smadja. The success of this thesis project leads to the development of a software for automated diagnosis of stroke MRI data. The main interest of Jonathan Kobolds research is semantic segmentation of medical imaging data with recurrent convolutional networks.

Date: 05/03/2020, 14h
Lieu: IBISC, site Pelvoux, amphithéâtre Bx30
Invitant: Vincent VIGNERON (MCF HDR Univ. Evry, équipe SIAM)
Organisatrice: Naïma AIT OUFROUKH-MAMMAR (MCF Univ. Evry, équipe SIAM)
Présentation de Jonathan Kobold (au format PDF)