English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Convolutional neural networks (CNNs) demonstrate promising accuracy in a widerange of applications. Among all layers in CNNs, convolution layers are themost computation-intensive and consume the most energy. As the maturity ofdevice and fabrication technology, 3D resistive random access memory (ReRAM)receives substantial attention for accelerating large vector-matrixmultiplication and convolution due to its high parallelism and energyefficiency benefits. However, implementing multi-channel convolution naively in3D ReRAM will either produce incorrect results or exploit only partialparallelism of 3D ReRAM. In this paper, we propose a 3D ReRAM-based convolutionaccelerator architecture, which efficiently maps multi-channel convolution tomonolithic 3D ReRAM. Our design has two key principles. First, we exploit theintertwined structure of 3D ReRAM to implement multi-channel convolution byusing a state-of-the-art convolution algorithm. Second, we propose a newapproach to efficiently implement negative weights by separating them fromnon-negative weights using configurable interconnects. Our evaluationdemonstrates that our mapping scheme in 16-layer 3D ReRAM achieves a speedup of5.79X, 927.81X, and 36.8X compared with a custom 2D ReRAM baseline andstate-of-the-art CPU and GPU. Our design also reduces energy consumption by2.12X, 1802.64X, and 114.1X compared with the same baseline.

Efficient Implementation of Multi-Channel Convolution in Monolithic 3D ReRAM Crossbar