Renesas Electronics has developed an AI accelerator that performs CNN (convolutional neural network) processing at high speeds and low power to move towards the next generation of Renesas embedded AI (e-AI), which will accelerate increased intelligence of endpoint devices.
The company says its first test chip featuring this accelerator has achieved the power efficiency of 8.8 TOPS/W.
Renesas developed the following three technologies for the new AI accelerator: a ternary-valued (-1, 0, 1) SRAM structure PIM technology that can perform large-scale CNN computations; an SRAM circuit to be applied with comparators that can read out memory data at low power; and a technology that prevents calculation errors due to process variations in the manufacturing.
https://www.renesas.com/us/en/about/press-center/news/2019/news20190613.html
