Neuromorphic computing architecture with dynamically accessible contexts

DWPI Title: Neuromorphic computing system for executing neural algorithm, has resource allocation system that identifies subgraphs of neural algorithm graph representative of operations of neural network algorithm
Abstract: Various technologies pertaining to allocating computing resources of a neuromorphic computing system are described herein. Subgraphs of a neural algorithm graph to be executed by the neuromorphic computing system are identified. The subgraphs are each executed by a group of neuron circuits serially. Output data generated by execution of the subgraphs are provided to the same or a second group of neuron circuits at a same time or with associated timing data indicative of a time at which the output data was generated. The same or second group of neuron circuits performs one or more processing operations based upon the output data.
Use: Neuromorphic computing system for executing neural algorithm.
Advantage: The neuromorphic computing system executes neural algorithms that have larger numbers of neurons than a number of available neuron circuits in the neural architecture.
Novelty: The system (100) has several synapse connections (110) connecting each of the neuron circuits (108) to other neuron circuit in the neuron circuits. A resource allocation system (104) allocates resources of the neuromorphic computing system to execute a neural network algorithm. The resource allocation system identifies several subgraphs of a neural algorithm graph representative of operations of the neural network algorithm. The resource allocation system assigns execution of the subgraphs to a first group of neuron circuits in the neuron circuits. The first group of neuron circuits is configured to provide the output data for each of the subgraphs to a second group of neuron circuits in the neuron circuits. The neuron circuits are configured such that the second group of neuron circuits receives the output data for each of the subgraphs at a same time. The second group of neuron circuits is configured to perform the processing operations based upon the output data.
Filed: 6/15/2017
Application Number: US15624271A
Tech ID: SD 14181.0
This invention was made with Government support under Contract No. DE-NA0003525 awarded by the United States Department of Energy/National Nuclear Security Administration. The Government has certain rights in the invention.
Data from Derwent World Patents Index, provided by Clarivate
All rights reserved. Republication or redistribution of Clarivate content, including by framing or similar means, is prohibited without the prior written consent of Clarivate. Clarivate and its logo, as well as all other trademarks used herein are trademarks of their respective owners and used under license.