Channel pruning is widely accepted to accelerate modern convolutional neural networks (CNNs). The resulting pruned model benefits from its immediate deployment on general-purpose software and hardware resources. However, its large pruning granularity, specifically at the unit of a convolution filter, often leads to undesirable accuracy drops due to the inflexibility of deciding how and where to introduce sparsity to the CNNs. In this paper, we propose REPrune, a novel channel pruning technique that emulates kernel pruning, fully exploiting the finer but structured granularity. REPrune identifies similar kernels within each channel using agglomerative clustering. Then, it selects filters that maximize the incorporation of kernel representatives while optimizing the maximum cluster coverage problem. By integrating with a simultaneous training-pruning paradigm, REPrune promotes efficient, progressive pruning throughout training CNNs, avoiding the conventional train-prune-finetune sequence. Experimental results highlight that REPrune performs better in computer vision tasks than existing methods, effectively achieving a balance between acceleration ratio and performance retention.

Depth-wise Separable Convolution (DSConv) has a powerful representation even with fewer parameters and computation, leading to its adoption by almost all of the state-of-the-art CNN models. DSConv models are already compact making it hard to apply pruning, and there are few previous pruning techniques that target depth-wise convolution (DW-conv). In this paper, we present Depth-wise Separable Convolution Pruning (DEPrune), a novel pruning method applied to both point-wise and depth-wise convolutions. DEPrune is optimized by analyzing the computation of DSConv on GPUs. DEPrune employs a fine-grained pruning approach, yet it achieves the structured sparsity typically absent in fine-grained pruning, enabling practical hardware acceleration. Moreover, this method maintains a high pruning ratio without causing any accuracy drop. We additionally represent techniques that further enhance DEPrune performance: 1) balanced workload tuning (BWT), and 2) hardware-aware sparsity recalibration (HSR). Experiment results show that DEPrune achieves up to $3.74\times$ practical speedup in DSConv inference on GPUs while maintaining the accuracy of EfficientNet-B0 on ImageNet.

Pruning has been an effective solution to reduce the number of computations and the memory requirement in deep learning. The pruning unit plays an important role in exploiting the GPU resources efficiently. The filter is proposed as a simple pruning unit of structured pruning. However, since the filter is quite large as pruning unit, the accuracy drop is considerable with a high pruning ratio. GPU rearranges the weight and input tensors into tiles (blocks) for efficient computation. To fully utilize GPU resources, this tile structure should be considered, which is the goal of block pruning. However, previous block pruning prunes both row vectors and column vectors. Pruning of row vectors in a tile corresponds to filter pruning, and it also interferes with column-wise block pruning of the following layer. In contrast, column vectors are much smaller than row vectors and can achieve lower accuracy drop. Additionally, if the pruning ratio for each tile is different, GPU utilization can be limited by imbalanced workloads by irregular-sized blocks. The same pruning ratio for the weight tiles processed in parallel enables the actual inference process to fully utilize the resources without idle time. This paper proposes balanced column-wise block pruning, named BCBP, to satisfy two conditions: the column-wise minimal size of the pruning unit and balanced workloads. We demonstrate that BCBP is superior to previous pruning methods through comprehensive experiments.

Dilated convolution is used to achieve wide receptive fields in computer vision algorithms such as image segmentation and denoising. Unlike the strided convolution, dilated convolution maintains the resolution of the output feature map same as the input feature map. Thus, the computational complexity can be increased to configure the convolutional neural network (CNN) architecture with the dilated convolutional layer. However, the complexity accordingly introduces additional computation delay and it is strongly required to have a proper way to lessen the computation delay of the dilated convolution. In this paper, we propose the dilated-Winograd convolution to reduce the computational complexity of the dilated convolution. By using the Winograd transform with a dilation rate, the number of pixels in the tile is effectively reduced. The proposed acceleration methods result in an average speedup of 2.043 and 1.456 with dilation rate of 2 and 4 compared to the state-of-the-art implementation.