![]()
ROHM Co., Ltd. has developed an ultra-low-power SoC with an on-device AI accelerator that enables real-time failure prediction in electronic devices equipped with motors and sensors. | Edge Impulse (a Qualcomm company)
ROHM Co., Ltd. has developed an ultra-low-power SoC with an on-device AI accelerator that enables real-time failure prediction in electronic devices equipped with motors and sensors.www.linkedin.com
Competizione’ ? Sounds similar but have not delved in …..
View attachment 23201
Rohm have a NN which uses magnetic MemRistors:
US2022172034A1 NEURAL NETWORK
A neural network includes first electrode lines in parallel, second electrode lines in parallel, a ferroelectric layer, neuron circuits, a first direction control circuit, and a second direction control circuit. The second electrode lines extend in a direction different from the first electrode lines. The ferroelectric layer is arranged between the first electrode lines and the second electrode lines. The neuron circuits are provided in the first electrode lines, respectively. The first direction control circuit is connected between the neuron circuits and the first electrode lines. The second direction control circuit is connected between the neuron circuits and the second electrode lines. The first electrode lines and the second electrode lines are capacitively coupled to form synapse devices at intersections in a plan view, each of the intersections being a portion where a first electrode line and a second electrode line intersect with each other.
[0005] When the synapse devices are implemented by variable resistance elements as in WO2019/078367 and Japanese Patent Laying-Open No. 2019-179499, a current should steadily be fed to the synapse devices in storing (training) and reading (retrieving) data. Therefore, in particular when the number of synapse devices increases for high integration, power consumption may increase.
[0006] The present disclosure was made to solve such a problem, and an object thereof is to lower power consumption in a hardware neural network.
Last edited: