LFD Book Forum

LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   General Discussion of Machine Learning (http://book.caltech.edu/bookforum/forumdisplay.php?f=105)
-   -   Types of Machine Learning (http://book.caltech.edu/bookforum/showthread.php?t=834)

cjohnson 07-09-2012 10:16 AM

Types of Machine Learning
 
In addition to perceptrons and neural networks, I have read about particle swarms and genetic algorithms. Will these also be addressed as well, or are they outside the realm of this course?

magdon 07-09-2012 07:02 PM

Re: Types of Machine Learning
 
perceptrons and neural networks (a generalization of the perceptron) are ways to formulate the hypothesis set \cal H that represents your target function f. There is also a second ingredient in learning, namely how you find a member of your hypothesis set g\in\cal H that fits your data. Particle swarms are a particular way of finding such a g and genetic algorithms also correspond to a related way to find your g when your g is obtained as the combination of simple functions using simple combination rules. In general, the way to find g falls under the topic of optimization, and so particle swarms and genetic algorithms are two ways of finding fits to the data that have "high fitness".

This course does not consider these particular ways to optimize (i.e. fit the data to produce g), but we do consider other optimization techniques based on different ideas.


Quote:

Originally Posted by cjohnson (Post 3358)
In addition to perceptrons and neural networks, I have read about particle swarms and genetic algorithms. Will these also be addressed as well, or are they outside the realm of this course?



All times are GMT -7. The time now is 04:04 AM.

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
The contents of this forum are to be used ONLY by readers of the Learning From Data book by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, and Hsuan-Tien Lin, and participants in the Learning From Data MOOC by Yaser S. Abu-Mostafa. No part of these contents is to be communicated or made accessible to ANY other person or entity.