English

LogicSparse: Enabling Engine-Free Unstructured Sparsity for Quantised Deep-learning Accelerators

Hardware Architecture 2025-11-06 v1

Abstract

FPGAs have been shown to be a promising platform for deploying Quantised Neural Networks (QNNs) with high-speed, low-latency, and energy-efficient inference. However, the complexity of modern deep-learning models limits the performance on resource-constrained edge devices. While quantisation and pruning alleviate these challenges, unstructured sparsity remains underexploited due to irregular memory access. This work introduces a framework that embeds unstructured sparsity into dataflow accelerators, eliminating the need for dedicated sparse engines and preserving parallelism. A hardware-aware pruning strategy is introduced to improve efficiency and design flow further. On LeNet-5, the framework attains 51.6 x compression and 1.23 x throughput improvement using only 5.12% of LUTs, effectively exploiting unstructured sparsity for QNN acceleration.

Keywords

Cite

@article{arxiv.2511.03079,
  title  = {LogicSparse: Enabling Engine-Free Unstructured Sparsity for Quantised Deep-learning Accelerators},
  author = {Changhong Li and Biswajit Basu and Shreejith Shanker},
  journal= {arXiv preprint arXiv:2511.03079},
  year   = {2025}
}

Comments

Accepted by ICFPT 2025

R2 v1 2026-07-01T07:22:11.233Z