International business machines corporation (20240346296). LOW-LATENCY TIME-ENCODED SPIKING NEURAL NETWORK simplified abstract

LOW-LATENCY TIME-ENCODED SPIKING NEURAL NETWORK

Organization Name

international business machines corporation

Inventor(s)

Giovanni Cherubini of Rueschlikon (CH)

Marcel A. Kossel of Reichenburg (CH)

LOW-LATENCY TIME-ENCODED SPIKING NEURAL NETWORK - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240346296 titled 'LOW-LATENCY TIME-ENCODED SPIKING NEURAL NETWORK

Simplified Explanation: The patent application describes a method for executing a time-encoded spiking neural network (TSNN) using an electronic circuit that connects pairs of neurons in the network. Signals are sent in parallel across channels between sender and receiver neurons, encoding subcycle timing information about spike timing relative to time steps.

• Configuring an electronic circuit to connect pairs of neurons in a time-encoded spiking neural network (TSNN).
• Operating the electronic circuit at an actual clock rate corresponding to time steps to send signals in parallel across channels.
• Signals encode subcycle timing information about spike timing relative to time steps.
• Emulating the execution of the TSNN at an effective clock rate by multiplying the actual clock rate by a latency reduction factor (LRF).

Potential Applications: 1. Neuromorphic computing 2. Artificial intelligence 3. Robotics 4. Brain-computer interfaces

Problems Solved: 1. Efficient execution of time-encoded spiking neural networks 2. Reduction of latency in neural network operations 3. Emulation of biological neural networks in electronic circuits

Benefits: 1. Improved performance in neural network computations 2. Faster processing of time-encoded information 3. Enhanced accuracy in spike timing simulations

Commercial Applications: The technology could be utilized in industries such as: 1. Healthcare for medical diagnostics 2. Automotive for autonomous driving systems 3. Aerospace for flight control systems 4. Telecommunications for signal processing

Prior Art: Prior research in neuromorphic computing and spiking neural networks may provide insights into similar methods of emulating biological neural networks in electronic circuits.

Frequently Updated Research: Researchers are constantly exploring advancements in neuromorphic computing and spiking neural networks to improve the efficiency and accuracy of neural network simulations.

Questions about Time-Encoded Spiking Neural Networks: 1. How does the latency reduction factor (LRF) impact the performance of the electronic circuit in emulating the TSNN? 2. What are the potential limitations of using parallel channels for signaling in the electronic circuit?

Original Abstract Submitted

a method of executing a time-encoded spiking neural network (tsnn) that includes configuring an electronic circuit connecting pairs of neurons of the tsnn, wherein each pair of the pairs of neurons connects a sender neuron to a receiver neuron through parallel channels, and operating the electronic circuit at an actual clock rate corresponding to actual time steps, for the electronic circuit to perform signaling over said each pair at each time step of the actual time steps by sending signals in parallel across the parallel channels. sent signals encode subcycle timing information about a timing of spikes relative to subcycle time steps, a unit duration that corresponds to a duration of said each time step divided by a latency reduction factor (lrf), for the operated electronic circuit to emulate an execution of the tsnn at an effective clock rate corresponding to the actual clock rate multiplied by the lrf.