<ul><li>Continual learning techniques employ simple replay sample selection processes and use them during subsequent tasks.</li><li>This paper proposes a label-free replay buffer and introduces cluster preservation loss in order to maintain essential information from previously encountered tasks while adapting to new tasks.</li><li>The method includes 'push-away' and 'pull-toward' mechanisms to retain previously learned information and facilitate adaptation to new classes or domain shifts.</li><li>Experimental results on various benchmarks show that the label-free replay-based technique outperforms state-of-the-art continual learning methods and even surpasses offline learning in some cases.</li></ul>

Prototype-Based Continual Learning with Label-free Replay Buffer and Cluster Preservation Loss

Discover more