Systems, methods, and computer devices for transistor-based neural networks

• Main Authors: LEVINSON REUVEN, GOSSEN, JOHN, L
• Format: Patent
• Language: Chinese; English
• Published: 27.02.2024
• Subjects: BASIC ELECTRONIC CIRCUITRY; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRICITY; IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; INFORMATION STORAGE; PHYSICS; RESONATORS; STATIC STORES

Computer systems, methods, and devices for operating an artificial neural network are provided. The system includes neurons. The neuron comprises: a plurality of synapses comprising a charge trapping transistor for processing an input signal; an accumulation block for receiving a drain current from the plurality of synapses, the drain current being generated as an output of a multiplication operation from the plurality of synapses, the drain current calculating a voltage amount multiplied by time; a capacitor for accumulating charge from the drain current to serve as a short-term memory of an accumulated signal; a discharge pulse generator for generating an output signal by discharging the accumulated charge during a discharge period; and a comparator for comparing the input voltage with a reference voltage. The comparator generates a first output if the input voltage is higher than a reference voltage, and generates a second output if the input voltage is lower than the reference voltage. 提供了用于操作人工神经网络的计算机系统