Introduction to Pattern Recognition. Pattern Recognition.

19
Introduction to Pattern Recognition

Transcript of Introduction to Pattern Recognition. Pattern Recognition.

Page 1: Introduction to Pattern Recognition. Pattern Recognition.

Introduction to Pattern Recognition

Page 2: Introduction to Pattern Recognition. Pattern Recognition.

Pattern Recognition

Page 3: Introduction to Pattern Recognition. Pattern Recognition.

What is pattern recognition?

A pattern is an object, process or event that can be given a name.

A pattern class (or category) is a set of patterns sharing common attributes and usually originating from the same source.

During recognition (or classification) given objects are assigned to prescribed classes.

A classifier is a machine which performs classification.

“The assignment of a physical object or event to one of several prespecified categeries” -- Duda & Hart

Page 4: Introduction to Pattern Recognition. Pattern Recognition.

Examples of applications

Page 5: Introduction to Pattern Recognition. Pattern Recognition.
Page 6: Introduction to Pattern Recognition. Pattern Recognition.
Page 7: Introduction to Pattern Recognition. Pattern Recognition.
Page 8: Introduction to Pattern Recognition. Pattern Recognition.
Page 9: Introduction to Pattern Recognition. Pattern Recognition.
Page 10: Introduction to Pattern Recognition. Pattern Recognition.
Page 11: Introduction to Pattern Recognition. Pattern Recognition.
Page 12: Introduction to Pattern Recognition. Pattern Recognition.
Page 13: Introduction to Pattern Recognition. Pattern Recognition.
Page 14: Introduction to Pattern Recognition. Pattern Recognition.
Page 15: Introduction to Pattern Recognition. Pattern Recognition.
Page 16: Introduction to Pattern Recognition. Pattern Recognition.
Page 17: Introduction to Pattern Recognition. Pattern Recognition.
Page 18: Introduction to Pattern Recognition. Pattern Recognition.

Overfitting and underfitting

Problem: how rich class of classifications q(x;θ) to use.

underfitting overfittinggood fit

Problem of generalization: a small emprical risk Remp does not imply small true expected risk R.

Page 19: Introduction to Pattern Recognition. Pattern Recognition.