Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
9 views6 pages

Lecture2 s12

The document discusses the concept of learning in computer programs, defining it as the improvement of performance at specific tasks through experience. A well-posed learning problem requires the identification of three key features: the class of tasks, the performance measure, and the source of experience. Examples provided include recognizing handwritten words, playing checkers, and classifying patients as healthy or sick, each illustrating the components of a learning problem.

Uploaded by

ghashian135
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
9 views6 pages

Lecture2 s12

The document discusses the concept of learning in computer programs, defining it as the improvement of performance at specific tasks through experience. A well-posed learning problem requires the identification of three key features: the class of tasks, the performance measure, and the source of experience. Examples provided include recognizing handwritten words, playing checkers, and classifying patients as healthy or sick, each illustrating the components of a learning problem.

Uploaded by

ghashian135
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 6

Well Posed Learning Problems

Lecture # 2
21st February, 2012

(Chapter 1)
What is learning
 Any computer program that improves its performance at some
task through experience

 A computer program is said to learn from experience E with


respect to some class of tasks T and performance measure P, if its
performance at tasks in T, as measured by P, improves with
experience E

Lecture #2 2
Well Posed Learning Problem
 To have a well defined learning problem, we must identify 3
features
 The class of tasks
 The measure of performance to be improved
 The source of experience

Lecture #2 3
Well Posed Learning Problem - Example

 Task T : Recognizing handwritten words

 Performance measure P : Percentage of correctly classified


words

 Training experience E : A database of handwritten words with


known classifications

Lecture #2 4
Well Posed Learning Problem - Example

 Task T : Playing checkers

 Performance measure P : Percentage of games won against


opponents

 Training experience E : Playing practice games against itself

Lecture #2 5
Well Posed Learning Problem - Example

 Task T : Classify patient as healthy or sick

 Performance measure P : Percentage of correctly classified


patients

 Training experience E : Classified patient records e.g


<<T1, BP1>, H>, <<T2,
BP1>, S>

 Learning: Find the target function V i.e. acquire the concepts H and
S from a subset of all possible experience
Lecture #2 6

You might also like