Transcription of CSC 411: Lecture 06: Decision Trees
{{id}} {{{paragraph}}}
Example: air traffic controller
CSC 411: Lecture 06: Decision Trees
csc 411 : Lecture 06: Decision TreesRichard Zemel, Raquel Urtasun and Sanja FidlerUniversity of TorontoZemel, Urtasun, Fidler (UofT) csc 411 : 06- Decision Trees1 / 39 TodayDecision TreesIentropyIinformation gainZemel, Urtasun, Fidler (UofT) csc 411 : 06- Decision Trees2 / 39 Another Classification IdeaWe learned about linear classification ( , logistic regression), and nearestneighbors. Any other idea?Pick an attribute, do a simple testConditioned on a choice, pick another attribute, do another testIn the leaves, assign a class with majority voteDo other branches as wellZemel, Urtasun, Fidler (UofT) csc 411 : 06- Decision Trees3 / 39 Another Classification IdeaGives axes aligned Decision boundariesZemel, Urtasun, Fidler (UofT) csc 411 : 06- Decision Trees4 / 39 Decision tree : ExampleYes No Yes No Yes No Zemel, Urtasun, Fidler (UofT) csc 411 : 06- Decision Trees5 / 39 Decision tree : ClassificationZemel, Urtasun, Fidler (UofT) csc 411 : 06- Decision Trees6 / 39 Example with Di
CSC 411: Lecture 06: Decision Trees Richard Zemel, Raquel Urtasun and Sanja Fidler University of Toronto Zemel, Urtasun, Fidler (UofT) CSC 411: 06-Decision Trees 1 / 39
Loading..
Tags:
Information
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document: