Course Website: [ Ссылка ]
Learning outcome:
Machine learning studies the design of models and training algorithms in order to learn how to solve tasks from data. Whereas historically machine learning has concentrated primarily on static predefined training datasets and respective test scenarios, recent advances also take into account the fact that the world is constantly evolving. In this course, we will go beyond the train-validate-test phase and explore modern approaches to machines that can learn continually. In addition to a comprehensive overview of the breath of factors to consider in continual learning, the course will delve into techniques that span mitigation of forgetting across multiple tasks, selection of new data in continuous training, dynamic model architectures, and robustness with respect to unexpected data inputs.
Course content:
The course is structured to provide a comprehensive overview of the many facets involved in design, training, and evaluation of continually evolving systems. Respectively explored topics include:
* Introduction and motivation to learning continually
* From domain adaptation and transfer to continual learning
* Alleviating catastrophic forgetting: methodologies and examples
* Active learning: selecting future data
* Modular and dynamic architectures
* Curriculum learning
* Closed and open world assumptions
* Continual learning benchmarks and metrics
* Learning to learn: a meta-learning perspective
* Software developments for continual learning
* Open-ended research questions and applications
If you like this course, feel free to subscribe to the ContinualAI Youtube channel: [ Ссылка ]
Join the ContinualAI community: [ Ссылка ]
Visit Hessian.AI: [ Ссылка ]
