Progressive Voronoi Diagram Subdivision: Towards A Holistic Geometric Framework for Exemplar-free Class-Incremental Learning. (arXiv:2207.14202v1 [cs.CV])

Exemplar-free Class-incremental Learning (CIL) is a challenging problem
because rehearsing data from previous phases is strictly prohibited, causing
catastrophic forgetting of Deep Neural Networks (DNNs). In this paper, we
present iVoro, a holistic framework for CIL, derived from computational
geometry. We found Voronoi Diagram (VD), a classical model for space
subdivision, is especially powerful for solving the CIL problem, because VD
itself can be constructed favorably in an incremental manner -- the newly added
sites (classes) will only affect the …

arxiv cv framework free incremental learning voronoi