Difference between revisions of "Inductive bias"

From Lesswrongwiki
Jump to: navigation, search
(fleshed out)
m
Line 8: Line 8:
 
|year = 1980
 
|year = 1980
 
|title = The need for biases in learning generalizations
 
|title = The need for biases in learning generalizations
|format = PDF
 
 
|url = http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.120.4179
 
|url = http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.120.4179
 
}}</ref>}}
 
}}</ref>}}
  
"Inductive bias" refers to your suspicion that if the sun has risen for the last billion days in a row, then it may rise tomorrow as well.  Since it is [[logically possible]] that the laws of physics will arbitrarily cease to work and that the sun will *not* rise tomorrow, coming to this conclusion requires an inductively biased [[prior]].
+
'''Inductive bias''' refers to your suspicion that if the sun has risen for the last billion days in a row, then it may rise tomorrow as well.  Since it is [[logically possible]] that the laws of physics will arbitrarily cease to work and that the sun will ''not'' rise tomorrow, coming to this conclusion requires an inductively biased [[prior]].
  
 
This sort of bias is not a bad thing - without "inductive bias" you can't draw any conclusion at all from the data.  It's just a different technical meaning attached to the same word.
 
This sort of bias is not a bad thing - without "inductive bias" you can't draw any conclusion at all from the data.  It's just a different technical meaning attached to the same word.
  
===Primary article===
+
==Primary post==
 +
 
 
*[http://lesswrong.com/lw/hg/inductive_bias/ "Inductive Bias"] by [[Eliezer Yudkowsky]]
 
*[http://lesswrong.com/lw/hg/inductive_bias/ "Inductive Bias"] by [[Eliezer Yudkowsky]]
  
===See also===
+
==Footnotes==
 +
<references/>
 +
 
 +
==See also==
 +
 
 
*[[Prior distribution]]
 
*[[Prior distribution]]
 
*[[Statistical bias]]
 
*[[Statistical bias]]
 
*[[Cognitive bias]]
 
*[[Cognitive bias]]
  
===Footnotes===
+
[[Category:Learning]]
<references/>
 

Revision as of 00:32, 27 June 2009

Smallwikipedialogo.png
Wikipedia has an article about


The inductive bias of a learning algorithm is the set of assumptions that the learner uses to predict outputs given inputs that it has not encountered

Tom Mitchell[1]

Inductive bias refers to your suspicion that if the sun has risen for the last billion days in a row, then it may rise tomorrow as well. Since it is logically possible that the laws of physics will arbitrarily cease to work and that the sun will not rise tomorrow, coming to this conclusion requires an inductively biased prior.

This sort of bias is not a bad thing - without "inductive bias" you can't draw any conclusion at all from the data. It's just a different technical meaning attached to the same word.

Primary post

Footnotes

See also