APPARATUS AND METHOD WITH ACCELERATING ARTIFICIAL NEURAL NETWORK: abstract simplified (18296165)
- This abstract for appeared for patent application number 18296165 Titled 'APPARATUS AND METHOD WITH ACCELERATING ARTIFICIAL NEURAL NETWORK'
Simplified Explanation
The abstract describes a computer processor apparatus that performs a specific type of mathematical operation called a Winograd convolution. This operation involves transforming input feature maps and kernels, multiplying them together, and then performing an inverse transform to generate output feature maps. The apparatus includes modules for each step of this process.
Original Abstract Submitted
A processor-implemented apparatus includes a forward transform module configured to transform input feature maps (IFMs) by performing a forward transform operation in a Winograd convolution (WinConv) domain, multiply and accumulate array (MAA) units configured to multiply the transformed IFMs by transformed kernels and perform a first inverse transform operation based on results of the multiplying, and an inverse transform module configured to generate output feature maps (OFMs) based on a result of the first inverse transform operation.