Recent Publications

PLOP: Learning without Forgetting for Continual Semantic Segmentation

Continual Semantic Segmentation with Local POD, a multi-scale distillation scheme preserving long- and short-range spatial relationships at feature level, and an uncertainty-based pseudo-labeling handling background shift.

Insights from the Future for Continual Learning

Instead of optimizing the past and current tasks, we train a model to be good on all tasks, future included. The incorporation of future tasks with ZeroShot allows us to gain insights useful to minimize interference with old and …

PODNet: Pooled Outputs Distillation for Small-Tasks Incremental Learning

We apply a novel distillation loss over all ConvNet’s intermediary features maps to reduce Catastrophic Forgetting in a large amount of small tasks.



A PyTorch library providing loaders for Continual Learning


Collection of Continual Learning paper implementations


Continual Learning   PDF
Continual Learning   PDF
PODNet: Small Task Incremental Learning   PDF   Video
Small Task Incremental Learning   PDF   Video
Incremental Learning   PDF
Lowshots learning   PDF
NATO Innovation Challenge   URL   Blog Post
Object Detection with Deep Learning   PDF
State of the FoodTech in France


Deep Learning
Deep Learning for Computer Vision   URL
Deep Learning for Computer Vision
C Programming
Unix / Shell / Algorithm theory
Java Programming
SQL (Postgres) Programming
Web Programming & Project Management
C++ Programming