MGiaD: Multigrid in all dimensions. Efficiency and robustness by coarsening in resolution and channel dimensions

TL;DR

This paper introduces MGiaD, a multigrid-inspired neural network architecture that coarsens in resolution and channel dimensions, achieving linear weight complexity, reducing overfitting, and improving image classification performance.

Contribution

It combines multigrid ideas across all dimensions to create a more efficient and robust neural network architecture with significantly fewer weights.

Findings

Reduces weight count compared to traditional architectures.

Improves overfitting resistance and generalization.

Outperforms ResNet on standard benchmarks.

Abstract

Current state-of-the-art deep neural networks for image classification are made up of 10 - 100 million learnable weights and are therefore inherently prone to overfitting. The complexity of the weight count can be seen as a function of the number of channels, the spatial extent of the input and the number of layers of the network. Due to the use of convolutional layers the scaling of weight complexity is usually linear with regards to the resolution dimensions, but remains quadratic with respect to the number of channels. Active research in recent years in terms of using multigrid inspired ideas in deep neural networks have shown that on one hand a significant number of weights can be saved by appropriate weight sharing and on the other that a hierarchical structure in the channel dimension can improve the weight complexity to linear. In this work, we combine these multigrid ideas to introduce a joint framework of multigrid inspired architectures, that exploit multigrid structures in all relevant dimensions to achieve linear weight complexity scaling and drastically reduced weight counts. Our experiments show that this structured reduction in weight count is able to reduce overfitting and thus shows improved performance over state-of-the-art ResNet architectures on typical image classification benchmarks at lower network complexity.