arXiv · 2611.17212
Theoretical and Empirical Analysis of Sparse Convolution
Abstract
Deep neural networks for image processing are increasingly deployed on resource-constrained edge devices such as drones, medical systems, and smartphones, where limited computational resources and memory bandwidth pose significant challenges for efficient inference. Pruning during training often produces highly sparse convolutional kernels, offering the potential to significantly reduce computation and memory requirements. However, exploiting this sparsity efficiently at runtime remains challenging due to irregular memory access patterns and workload imbalance across processing elements. In this work, we investigate the standard convolution with sparse weights and analyze its performance characteristics through roofline analysis, showing that a 3 x 3 sparse convolution achieves an arithmetic intensity of approximately 10 FLOP/byte compared to roughly 96 FLOP/byte for its dense counterpart. We propose techniques for an efficient sparse convolution at high sparsities. Preliminary results demonstrate that our approach achieves upto 1.65x speedup as compared to dense convolution on NVIDIA A100 and upto 1.52x on NVIDIA Jetson Nano (Tegra X1) , motivating further exploration of direct convolution as a practical path to efficient sparse inference on modern accelerators/edge devices.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Keshvi Tuteja, Yu-Hsiang Tsai, Gregor Olenik, Markus Götz, Achim Streit, Hartwig Anzt, Charlotte Debus. 2026-11-12. Theoretical and Empirical Analysis of Sparse Convolution. https://searcharxiv.com/abs/2611.17212
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.