http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-104662526-B
Outgoing Links
| Predicate | Object |
|---|---|
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-049 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F15-18 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G10L25-30 |
| filingDate | 2013-07-25-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| grantDate | 2017-12-19-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationDate | 2017-12-19-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-104662526-B |
| titleOfInvention | Apparatus and method for efficiently updating spiking neuron network |
| abstract | The efficient renewal of the connection in artificial neural network can be realized.The connection using linear cynapse dynamic process can be described using framework, wherein these characteristics of connection with stable equilibrium.Neuron in the network and the state of cynapse can be updated based on the input and output to/from neuron.In some implementations, these renewals can be realized at regular time intervals.In one or more are realized, network activity (for example, neuron output and/or input) can be based on, these renewals are desirably realized, further to reduce the calculated load associated with synapse turnover.These connection renewals can be resolved into multiple connection change components dependent on event, wherein these components can be used for description due to neuron inputs and caused by connect Changes of Plasticity.By using the connection change component dependent on event, connection renewal can be performed on the basis of each neuron, rather than renewal is performed on the basis of each connection. |
| priorityDate | 2012-07-27-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 15.