An energy efficient processor array and memory controller for accurate processing of convolutional neural network-based inference engines

Abstract Exploiting unstructured sparsity in the hardware accelerator of a Convolutional Neural Networks (CNNs) based inference can improve energy efficiency. However, it needs a complex controller for indexing and load-balancing. A controller for managing unstructured sparsity in Fully Connected (F...

ver descrição completa

Detalhes bibliográficos

Principais autores:	S. Deepika, V. Arunachalam
Formato:	Artigo
Idioma:	Inglês
Publicado em:	Nature Portfolio 2025-11-01
coleção:	Scientific Reports
Assuntos:	Convolutional neural network Dataflow Deep learning Energy-efficient accelerator Image classification Compression
Acesso em linha:	https://doi.org/10.1038/s41598-025-23303-5
Tags:	Adicionar Tag Sem tags, seja o primeiro a adicionar uma tag!