?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Pattern+Recognition+with+Slow+Feature+Analysis&rft.creator=Berkes%2C+Pietro&rft.subject=Computational+Neuroscience&rft.subject=Machine+Learning&rft.subject=Neural+Nets&rft.description=Slow+feature+analysis+(SFA)+is+a+new+unsupervised+algorithm+to+learn+nonlinear+functions+that+extract+slowly+varying+signals+out+of+the+input+data.++In+this+paper+we+describe+its+application+to+pattern+recognition.+In+this+context+in+order+to+be+slowly+varying+the+functions+learned+by+SFA+need+to+respond+similarly+to+the+patterns+belonging+to+the+same+class.+We+prove+that%2C+given+input+patterns+belonging+to+C+non-overlapping+classes+and+a+large+enough+function+space%2C+the+optimal+solution+consists+of+C-1+output+signals+that+are+constant+for+each+individual+class.+As+a+consequence%2C+their+output+provides+a+feature+space+suitable+to+perform+classification+with+simple+methods%2C+such+as+Gaussian+classifiers.+We+then+show+as+an+example+the+application+of+SFA+to+the+MNIST+handwritten+digits+database.+The+performance+of+SFA+is+comparable+to+that+of+other+established+algorithms.+Finally%2C+we+suggest+some+possible+extensions+to+the+proposed+method.+Our+approach+is+in+particular+attractive+because+for+a+given+input+signal+and+a+fixed+function+space+it+has+no+parameters%2C+it+is+easy+to+implement+and+apply%2C+and+it+has+low+memory+requirements+and+high+speed+during+recognition.+SFA+finds+the+global+solution+(within+the+considered+function+space)+in+a+single+iteration+without+convergence+issues.+Moreover%2C+the+proposed+method+is+completely+problem-independent.%0A&rft.date=2005-02&rft.type=Preprint&rft.type=NonPeerReviewed&rft.format=application%2Fpdf&rft.identifier=http%3A%2F%2Fcogprints.org%2F4104%2F1%2FBerkes2005a-preprint.pdf&rft.format=application%2Fpostscript&rft.identifier=http%3A%2F%2Fcogprints.org%2F4104%2F2%2FBerkes2005a-preprint.ps&rft.identifier=++Berkes%2C+Pietro++(2005)+Pattern+Recognition+with+Slow+Feature+Analysis.++%5BPreprint%5D+++++&rft.relation=http%3A%2F%2Fcogprints.org%2F4104%2F