Large and performant neural networks are often overparameterized and can be drastically reduced in size and complexity thanks to pruning. Pruning is a group of methods, which seeks to remove redundant or unnecessary weights or groups of weights in a network. These techniques allow the creation of lightweight networks, which are particularly critical in embedded or mobile applications. In this paper, we devise an alternative pruning method that allows extracting effective subnetworks from larger untrained ones. Our method is stochastic and extracts subnetworks by exploring different topologies which are sampled using Gumbel Softmax. The latter is also used to train probability distributions which measure the relevance of weights in the sampled topologies. The resulting subnetworks are further enhanced using a highly efficient rescaling mechanism that reduces training time and improves performance. Extensive experiments conducted on CIFAR show the outperformance of our subnetwork extraction method against the related work.

翻译：大型高性能神经网络常存在过参数化问题，可通过剪枝大幅降低其尺寸与复杂度。剪枝是一类从网络中移除冗余或不必要权重（或权重组）的方法。这类技术可生成轻量级网络，在嵌入式或移动应用中尤为关键。本文提出一种替代性剪枝方法，能够从未经训练的大型网络中提取有效子网络。该方法采用随机策略，通过Gumbel-Softmax采样探索不同拓扑结构，并利用其训练概率分布以衡量各采样拓扑中权重的相关性。进一步地，通过高效的重缩放机制增强提取的子网络，该机制可缩减训练时间并提升性能。在CIFAR数据集上的大量实验表明，本文提出的子网络提取方法性能优于相关现有工作。