Rehearsal-Free Continual Learning Over Small Non-I.I.D. Batches

Vincenzo Lomonaco, Davide Maltoni, Lorenzo Pellegrini; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2020, pp. 246-247

Robotic vision is a field where continual learning can play a significant role. An embodied agent operating in a complex environment subject to frequent and unpredictable changes is required to learn and adapt continuously. In the context ofobject recognition, for example, a robot should be able to learn (without forgetting) objects ofnever before seen classes as well as improving its recognition capabilities as new instances of already known classes are discovered. Ideally, continual learning should be triggered by the availability ofshort videos ofsingle objects and performed on-line on on-board hardware with fine-grained updates. In this paper, we introduce a novel continual learning protocol based on the CORe50 benchmark and propose two rehearsal-free continual learning techniques, CWR* and AR1*, that can learn effectively even in the challenging case ofnearly 400 small non-i.i.d. incremental batches. In particular, our experiments show that AR1* can outperform other state-of-the-art rehearsal-free techniques by more than 15% accuracy in some cases, with a very light and constant computational and memory overhead across training batches.