?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Learning+as+Extraction+of+Low-Dimensional+Representations&rft.creator=Edelman%2C+Shimon&rft.creator=Intrator%2C+Nathan&rft.subject=Cognitive+Psychology&rft.description=Psychophysical+findings+accumulated+over+the+past+several+decades+indicate+that+perceptual+tasks+such+as+similarity+judgment+tend+to+be+performed+on+a+low-dimensional+representation+of+the+sensory+data.+Low+dimensionality+is+especially+important+for+learning%2C+as+the+number+of+examples+required+for+attaining+a+given+level+of+performance+grows+exponentially+with+the+dimensionality+of+the+underlying+representation+space.+In+this+chapter%2C+we+argue+that%2C+whereas+many+perceptual+problems+are+tractable+precisely+because+their+intrinsic+dimensionality+is+low%2C+the+raw+dimensionality+of+the+sensory+data+is+normally+high%2C+and+must+be+reduced+by+a+nontrivial+computational+process%2C+which%2C+in+itself%2C+may+involve+learning.+Following+a+survey+of+computational+techniques+for+dimensionality+reduction%2C+we+show+that+it+is+possible+to+learn+a+low-dimensional+representation+that+captures+the+intrinsic+low-dimensional+nature+of+certain+classes+of+visual+objects%2C+thereby+facilitating+further+learning+of+tasks+involving+those+objects.&rft.date=1997&rft.type=Preprint&rft.type=PeerReviewed&rft.format=application%2Fpostscript&rft.identifier=http%3A%2F%2Fcogprints.org%2F562%2F2%2F199710004.ps&rft.identifier=++Edelman%2C+Shimon+and+Intrator%2C+Nathan++(1997)+Learning+as+Extraction+of+Low-Dimensional+Representations.++%5BPreprint%5D+++++&rft.relation=http%3A%2F%2Fcogprints.org%2F562%2F