Machine Learning and Data Mining Lecture Notes

← Back to document page

Machine Learning and Data MiningLecture NotesCSC 411/D11Computer Science DepartmentUniversity of TorontoVersion: February 6, 2012Copyrightc 2010 Aaron Hertzmann and David FleetCSC 411 / CSC D11CONTENTSContentsConventions and Notationiv1Introduction to Machine of Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . simple problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..22Linear 1D case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . inputs . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . outputs.

it works on future test data. When a model fits training data well, but performs poorly on test data, we say that the model has overfit the training data; i.e., the model has fit properties of the input that are not particularly relevant to the task at hand (e.g., Figures 1 (top row and bottom left)). Such properties are refered to as noise.

  Lecture, Notes, Tests, Machine, Learning, Lecture notes, Machine learning