Continual Learning

Continual Learning

Internship Description

Continual learning aims to learn new tasks without forgetting previously learned ones. This is especially challenging when one cannot access data from previous tasks and when the model has a fixed capacity. In this project, the goal is to develop and improve the capability of the machine learning methods not to forget older concepts as time passes.

References

[1]  Arslan Chaudhry, Marc'Aurelio Ranzato, Marcus Rohrbach, Mohamed Elhoseiny, Efficient Lifelong Learning with A-GEM, ICLR, 2019

[2]  Mohamed Elhoseiny,Francesca Babiloni, Rahaf Aljundi, Manohar Paluri, Marcus Rohrbach, Tinne Tuytelaars, Exploring the Challenges towards Lifelong Fact Learning, ACCV 2018

https://arxiv.org/abs/1711.09601

[3]  Rahaf Aljundi, Francesca Babiloni, Mohamed Elhoseiny, Marcus Rohrbach, Tinne Tuytelaars, Memory Aware Synapses: Learning what (not) to forget, ECCV 2018

https://arxiv.org/abs/1711.09601

[4]Sayna Ebrahimi, Mohamed Elhoseiny, Trevor Darrell, Marcus Rohrbach Uncertainty-guided Continual Learning with Bayesian Neural Networks https://arxiv.org/abs/1906.02425

For more references, you may visit https://nips.cc/Conferences/2018/Schedule?showEvent=10910 https://icml.cc/Conferences/2019/Schedule?showEvent=3528​​​

Deliverables/Expectations

​Develop a working research prototype for a continual learning approach

1) students should learn about machine learning, deep learning, and the respective target application chosen for the internship.

2) students are expected to show capability to go from an idea to a working prototype; pushing the limits of what the state of the art can do.​

Faculty Name

Mohamed ELhoseiny

Field of Study

​Computer Vision and Machine Learning