Présentation en anglais

Réseaux profonds sous contraintes : au-delà de la descente de gradient

CONFÉRENCIER

Ismail Ben Ayed, professeur agrégé, École de technologie supérieure (ÉTS), 
Chaire de recherche ÉTS sur l'intelligence artificielle en imagerie médicale.

BIOGRAPHIE

Constrained Deep Networks: Beyond Gradient Descent

SPEAKER

Ismail Ben Ayed, Associate Professor, École de technologie supérieure (ÉTS),
ÉTS Research Chair on Artificial Intelligence in Medical Imaging.

ABSTRACT

Embedding priors and/or constraints on the outputs of deep networks has wide applicability in learning, vision and medical imaging. For instance, in weakly supervised learning, constraints can mitigate the lack of full and laborious annotations in dense prediction tasks, e.g., semantic segmentation. Also, adversarial robustness, which currently attracts substantial interest in the field, amounts to imposing constraints on network outputs. In this talk, I will discuss some recent developments in those research directions. A first part of the talk focuses on how to enforce various types of priors on weakly supervised convolutional neural networks (CNNs), which can leverage unlabeled data, guiding training with domain-specific knowledge. I will discuss several key technical aspects in the context of CNNs with partial or uncertain labels, including constrained optimization and popular Laplacian regularization. In the second part, I will discuss some state-of-the-art models for adversarial robustness. In both parts, I will emphasize how more attention should be paid to optimization methods, going beyond standard gradient descent. In particular, I will show how powerful discrete optimization techniques, e.g., alpha-expansion, can be very useful in imposing priors on CNNs, which promises to tackle a wide range of problems.  The talk includes various illustrations, applications and experimental results.

BIOGRAPHY

Ismail Ben Ayed is currently Associate Professor at ÉTS Montreal, where he holds a research Chair on Artificial Intelligence in Medical Imaging. His interests are in computer vision, optimization, machine learning and medical imaging. Ismail authored over 90 fully peer-reviewed papers, mostly published in the top venues of the field, along with 2 books and 7 US patents.  In the last 5 years, he gave over 20 invited talks, including 3 tutorials at flagship conferences (MICCAI’14, ISBI’16 and MICCAI’19). His research has been covered in several visible media outlets, such as Radio Canada (CBC), Quebec Science Magazine and Canal du Savoir. His team received several recent distinctions, such as MIDL’19 best paper runner-up award, several top-ranking positions in internationally visible contests (e.g., NeurIPS’18 adversarial vision challenge and MICCAI’17 iSeg Challenge), Medical Physics Editor’s choice, highly competitive FRQNT fellowships, and 6 oral presentations at prestigious conferences such as CVPR/ECCV/NeurIPS (3% acceptance rate), among other recognitions. Ismail served as Program Committee for MICCAI’15, MICCAI’17 and MICCAI’19, Program Chair for IEEE IPTA’17, and will serve as Program Chair for MIDL’20. Also, he serves regularly as reviewer for the main publications of the field, and received the outstanding reviewer award for CVPR’15.

Les séminaires scientifiques du CRIM, gratuits et ouverts à tous, sont donnés par des experts de renommée internationale, des collaborateurs universitaires, le personnel de R-D et les étudiants du CRIM. Au programme, des présentations conviviales sur les dernières avancées scientifiques et technologiques.

Conférence gratuite. Inscription requise.