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...

Fuld beskrivelse

Bibliografiske detaljer

Principais autores:	S. Deepika, V. Arunachalam
Format:	Artigo
Sprog:	Inglês
Udgivet:	Nature Portfolio 2025-11-01
Serier:	Scientific Reports
Fag:	Convolutional neural network Dataflow Deep learning Energy-efficient accelerator Image classification Compression
Online adgang:	https://doi.org/10.1038/s41598-025-23303-5
Tags:	Tilføj Tag Ingen Tags, Vær først til at tagge denne postø!