Search Results for author:
Found 2 papers, 1 papers with code
Incremental Task Learning with Incremental Rank Updates
Our method also offers better memory efficiency compared to episodic memory- and mask-based approaches.
Continual Learning via Low-Rank Network Updates
To update the network for a new task, we learn a low-rank (or rank-1) matrix and add that to the weights of every layer.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
