Nonparametric Teaching of Implicit Neural Representations

arXiv:2405.10531v1 Announce Type: new
Abstract: We investigate the learning of implicit neural representation (INR) using an overparameterized multilayer perceptron (MLP) via a novel nonparametric teaching perspective. The latter offers an efficient example selection framework for teaching nonparametrically defined (viz. non-closed-form) target functions, such as image functions defined by 2D grids of pixels. To address the costly training of INRs, we propose a paradigm called Implicit Neural Teaching (INT) that treats INR learning as a nonparametric teaching problem, where the given …

abstract arxiv cs.cv cs.lg example form framework functions image implicit neural representations mlp novel perceptron perspective pixels representation teaching type via viz