US Patent Application 17734221. METHOD FOR OPERATING NEURAL NETWORK simplified abstract
Contents
METHOD FOR OPERATING NEURAL NETWORK
Organization Name
Taiwan Semiconductor Manufacturing Company, Ltd.
Inventor(s)
Kea Tiong Tang of Hsinchu (TW)
METHOD FOR OPERATING NEURAL NETWORK - A simplified explanation of the abstract
This abstract first appeared for US patent application 17734221 titled 'METHOD FOR OPERATING NEURAL NETWORK
Simplified Explanation
The patent application describes a method for processing spike signals in a spiking neural network.
- Spike signals are received in an input layer of the network during multiple time steps.
- The number of spikes in each time step is counted.
- If the number of spikes in a time step exceeds a predetermined count value, the spike signals are weighted with synaptic weight values.
- Multiple synaptic signals are generated by applying the weighted spike signals.
- A membrane potential is generated by accumulating a certain number of synaptic signals based on their weight distribution.
- An output spike signal is generated based on the membrane potential.
Original Abstract Submitted
A method is provided and includes operations as below: receiving multiple spike signals in an input layer of a spiking neural network during multiple time steps; counting a corresponding number of spikes in the spike signals for each of the time steps; weighting, in response to the corresponding number of spikes in one of plurality of time steps being greater than a predetermined count value, the spike signals with multiple synaptic weight values to generate multiple synaptic signals; generating a membrane potential by accumulating a number N of the synaptic signals according to a weight distribution of the synaptic weight values; and generating an output spike signal according to the membrane potential.