Pages that link to "Item:Q2283825"

The following pages link to A faster algorithm for reducing the computational complexity of convolutional neural networks (Q2283825):

Displaying 6 items.

• Application of the residue number system to reduce hardware costs of the convolutional neural network implementation (Q1998097) (← links)
• Convolution accelerator designs using fast algorithms (Q2004889) (← links)
• A decomposable Winograd method for N-D convolution acceleration in video analysis (Q2054386) (← links)
• Pruning deep convolutional neural networks architectures with evolution strategy (Q2126266) (← links)
• Improved heuristics for low-latency implementations of linear layers (Q6080928) (← links)
• Method for Convolutional Neural Network Hardware Implementation Based on a Residue Number System (Q6486103) (← links)