http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114970836-A
Outgoing Links
| Predicate | Object |
|---|---|
| assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_7c037a5bb5d58b1748c02d930c8c2476 http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_d6a6f422b091ba12ea61d4adbf1b0e8e |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-048 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/H04J14-0202 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/H04B10-503 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-084 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/H04B10-50 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/H04J14-02 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 |
| filingDate | 2022-07-28-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_6bdd3b48ed4dffe9a97e5bb96f572d8e http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_d36ece7843178e5913d1cd09236389be http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_5776200401848d6a04eec22ebb45b260 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_66ac5866e752a9e40d9c8cb7bb48b582 |
| publicationDate | 2022-08-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-114970836-A |
| titleOfInvention | Reservoir neural network implementation method, system, electronic device and storage medium |
| abstract | The invention discloses a method for realizing a neural network of a water reservoir, which includes transmitting an optical signal through a laser, loading input information and feedback information corresponding to other wavelengths to the optical signal through a modulator, and combining with a wavelength division multiplexing unit to obtain wavelength division multiplexing The signal is input into the MZI array; the MZI array operates on the wavelength division multiplexing signal matrix, obtains the wavelength optical signal through the wavelength division multiplexing separation, converts the wavelength optical signal into an electrical signal through the detector, and applies an activation function to the electrical signal through the FPGA Obtain feedback information; loop iteration to obtain stable feedback information as the state of the reservoir node, and then input it to the initial fully connected neural network, and obtain the fully connected neural network after training through the cross entropy loss function; this method can improve the reservoir neural network. The computing speed is lower, the power is lower, and it has better fitting ability. The invention also provides a reservoir neural network realization system, electronic equipment and storage medium. |
| priorityDate | 2022-07-28-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 27.