Meta-Learning Sparse Implicit Neural Representations

TL;DR

This paper introduces a meta-learning approach combined with sparsity constraints to efficiently learn implicit neural representations, enabling quick adaptation to new signals with fewer parameters and less computation.

Contribution

It proposes a novel meta-learning method that leverages sparse neural networks for implicit representations, improving scalability and efficiency over dense models.

Findings

Meta-learned sparse models outperform dense models in loss reduction.

Sparse models adapt quickly to unseen signals.

Significant reduction in memory and computation requirements.

Abstract

Implicit neural representations are a promising new avenue of representing general signals by learning a continuous function that, parameterized as a neural network, maps the domain of a signal to its codomain; the mapping from spatial coordinates of an image to its pixel values, for example. Being capable of conveying fine details in a high dimensional signal, unboundedly of its domain, implicit neural representations ensure many advantages over conventional discrete representations. However, the current approach is difficult to scale for a large number of signals or a data set, since learning a neural representation -- which is parameter heavy by itself -- for each signal individually requires a lot of memory and computations. To address this issue, we propose to leverage a meta-learning approach in combination with network compression under a sparsity constraint, such that it renders a well-initialized sparse parameterization that evolves quickly to represent a set of unseen signals in the subsequent training. We empirically demonstrate that meta-learned sparse neural representations achieve a much smaller loss than dense meta-learned models with the same number of parameters, when trained to fit each signal using the same number of optimization steps.