18521665. NEURAL NETWORK DEVICE AND SYNAPTIC WEIGHT UPDATE METHOD simplified abstract (KABUSHIKI KAISHA TOSHIBA)
Contents
NEURAL NETWORK DEVICE AND SYNAPTIC WEIGHT UPDATE METHOD
Organization Name
Inventor(s)
Yoshifumi Nishi of Yokohama Kanagawa (JP)
Kumiko Nomura of Shinagawa Tokyo (JP)
Takao Marukame of Chuo Tokyo (JP)
Koichi Mizushima of Kamakura Kanagawa (JP)
NEURAL NETWORK DEVICE AND SYNAPTIC WEIGHT UPDATE METHOD - A simplified explanation of the abstract
This abstract first appeared for US patent application 18521665 titled 'NEURAL NETWORK DEVICE AND SYNAPTIC WEIGHT UPDATE METHOD
Simplified Explanation:
The neural network device described in the abstract consists of neuron circuits, synapse circuits, and random number circuits. The random number circuits generate random signals, which are used to update synaptic weights in the synapse circuits. These synapse circuits are grouped together based on the random signals they receive and the output signals they send to neuron circuits.
- Neuron circuits, synapse circuits, and random number circuits are key components of the neural network device.
- Random number circuits generate random signals used to update synaptic weights in synapse circuits.
- Synapse circuits are grouped based on the random signals they receive and the output signals they send to neuron circuits.
Potential Applications: This technology could be applied in various fields such as artificial intelligence, machine learning, robotics, and neuroscience research.
Problems Solved: This technology addresses the need for efficient and flexible neural network devices that can adapt and learn from random signals.
Benefits: The benefits of this technology include improved learning capabilities, adaptability, and flexibility in neural network systems.
Commercial Applications: Title: "Innovative Neural Network Device for Advanced AI Applications" This technology could be used in industries such as healthcare, finance, autonomous vehicles, and cybersecurity for advanced data processing and decision-making.
Prior Art: Researchers can explore prior art related to neural network devices, random signal processing, and synaptic weight updating algorithms in the field of artificial intelligence and neuroscience.
Frequently Updated Research: Researchers are constantly exploring new algorithms and techniques to enhance the performance and efficiency of neural network devices using random signals.
Questions about Neural Network Devices: 1. How do random number circuits contribute to the learning process in neural networks? 2. What are the potential limitations of using random signals in updating synaptic weights in neural network devices?
Original Abstract Submitted
A neural network device according to an embodiment includes a plurality of neuron circuits, a plurality of synapse circuits, and a plurality of random number circuits. Each of the random number circuits outputs a random signal. Each of the synapse circuits receives the random signal from one of the random number circuits and updates a synaptic weight with a probability generated on the basis of the received random signal. The synapse circuits are divided into synapse groups. Each of two or more synapse circuits belonging to a first synapse group receives the random signal output from a first random number circuit. Each of two or more synapse circuits outputting output signals to a first neuron circuit belongs to a synapse group differing from a synapse group, to which other synapse circuits outputting the output signal to the first neuron circuit, belong.
