CDNLive: Cadence intros fast neural network processor

Chris Rowen: neural network processing is big focus Vision processing and deep learning are the fastest growing applications areas for neural network processing. According to Chris Rowen, CTO for the IP Group at Cadence, speaking in Munich this week, advanced vision and deep learning is a big focus for Cadence. Neural network methodologies are used for running the new instructions and increased maths throughput needed for imaging and computer vision systems. “We are devoting intense efforts to discovering improved structures and training for neural networks, providing rich software environments for fast application development and offering breakthrough vision DSP architectures for embedded vision and learning deployment,” said Rowen. According to Cadence, the Vision P6 DSP quadruples multiply-accumulate (MAC) performance compared to the previous generation Vision P5 DSP. This is important for…


Link to Full Article: CDNLive: Cadence intros fast neural network processor