Progress & Compress: A scalable framework for continual learning

@inproceedings{Schwarz2018ProgressC,
  title={Progress & Compress: A scalable framework for continual learning},
  author={Jonathan Schwarz and Jelena Luketina and Wojciech Czarnecki and Agnieszka Grabska-Barwinska and Yee Whye Teh and Razvan Pascanu and Raia Hadsell},
  booktitle={ICML},
  year={2018}
}
We introduce a conceptually simple and scalable framework for continual learning domains where tasks are learned sequentially. Our method is constant in the number of parameters and is designed to preserve performance on previously encountered tasks while accelerating learning progress on subsequent problems. This is achieved by training a network with two components: A knowledge base, capable of solving previously encountered problems, which is connected to an active column that is employed to… CONTINUE READING

Similar Papers

Loading similar papers…