Transcription of Model Compression - Cornell University
{{id}} {{{paragraph}}}
ModelCompressionCristianBucil hundredsor thousandsof base-level classi , thespacerequiredto storethismany clas-si ers,andthetimerequiredto executethemat run-time,prohibitstheirusein applicationswheretestsetsarelarge( ),wherestoragespaceis ata premium( ),andwherecomputationalpower is limited( ).We present a method for\compressing" large ,complexensemblesin to smaller,fastermodels,usuallywith-outsign i cant lossin Subject [PatternRe-cognition]:Models{ :Algorithms,Experimentation,Measure-ment , Performance, :SupervisedLearning, a collectionof modelswhosepredictionsarecombinedby weightedaveragingor beenthefocusof signi cant research in thepastdecade,anda variety of ensemblemethods have knownensemblemethods includebagging[2],boosting[14],randomfor ests[3],Bayesianaveraging[9]andstacking[ 17].}
method introduced by Caruana et al. [5]. To achieve this, we take advantage of the well known property of arti cial neural nets, namely that they are universal approximators: givenenoughtraining data, anda large enoughhiddenlayer, a neural net can approximate any function to arbitrary pre-cision. Instead of training the neural net on the original
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document:
{{id}} {{{paragraph}}}