Predicate |
Object |
assignee |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_3ed37d724564735d73052c3c9c8e9e23 |
classificationCPCInventive |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-084 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 |
classificationIPCInventive |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N99-00 |
filingDate |
2017-02-06-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_59d539a4f16b0ca777336cd1bfe3910c http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_03a23c9e198b5de86b53f347becba2f5 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_275720e3ccf25b30c802d5d79ccd4b01 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_6302f31d46e9a09f0626a306a1bff17b |
publicationDate |
2018-10-25-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber |
DE-112017000670-T5 |
titleOfInvention |
Accelerate training of a deep neural network with inconsistent stochastic gradient decay |
abstract |
Aspects of the present disclosure describe techniques for training a convolutional neural network using an inconsistent stochastic gradient decay (ISGD) algorithm. A training effort to train batches, used by the ISGD algorithm, is dynamically adjusted according to a certain loss for a given training of a batch, which are classified into two sub-states - well-trained or under-trained. The ISGD algorithm provides more iterations for under-trained batches while reducing iterations for well-trained batches. |
priorityDate |
2016-02-05-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type |
http://data.epo.org/linked-data/def/patent/Publication |