Differentiable Transportation Pruning
This addresses the need for efficient neural network deployment on edge devices, representing a novel method for a known bottleneck rather than incremental work.
The paper tackles the problem of deploying deep neural networks on resource-constrained edge devices by proposing a novel pruning technique that provides precise control over output network size, achieving state-of-the-art performance across 3 datasets, 5 models, and various pruning ratios and sparsity budgets.
Deep learning algorithms are increasingly employed at the edge. However, edge devices are resource constrained and thus require efficient deployment of deep neural networks. Pruning methods are a key tool for edge deployment as they can improve storage, compute, memory bandwidth, and energy usage. In this paper we propose a novel accurate pruning technique that allows precise control over the output network size. Our method uses an efficient optimal transportation scheme which we make end-to-end differentiable and which automatically tunes the exploration-exploitation behavior of the algorithm to find accurate sparse sub-networks. We show that our method achieves state-of-the-art performance compared to previous pruning methods on 3 different datasets, using 5 different models, across a wide range of pruning ratios, and with two types of sparsity budgets and pruning granularities.