User Tools

Site Tools


Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
papers:tr-bottou-2012 [2012/09/12 08:13]
leonb
papers:tr-bottou-2012 [2018/05/01 10:22] (current)
leonb
Line 2: Line 2:
  
 //Abstract//: //Abstract//:
-Using the search engine ad placement problem as an example, we explain the +This work shows how to leverage causal inference to understand the behavior of complex learning systems interacting with their environment and predict the consequences of changes to the system. Such predictions allow both humans and algorithms to select changes that improve both the short-term and long-term performance of such systems. This work is illustrated by experiments carried out on the ad placement system associated with the Bing search engine.
-central role of causal inference for the design of learning systems interacting +
-with their environments. Thanks to importance sampling techniques, data +
-collected during randomized experiments gives precious cues to assist the +
-designer of such learning systems and useful signals to drive learning +
-algorithms. Thanks to a sharp distinction between the learning algorithms and +
-the extraction of the signals that drive them, these methods can be tailored +
-to causal models with different structures.  Thanks to mathematical +
-foundations shared with physics, these signals can describe the response of +
-the system when equilibrium conditions are reached.+
  
 <box 99% green> <box 99% green>
-Léon Bottou, Jonas Peters, Joaquin Quiñonero-Candela, Denis X. Charles, D. Max Chickering, Elon Portugualy, Dipankar Ray, Patrice Simard and Ed Snelson:  //**Counterfactual Reasoning and Learning Systems**//, arXiv:1209.2355, September 2012.+Léon Bottou, Jonas Peters, Joaquin Quiñonero-Candela, Denis X. Charles, D. Max Chickering, Elon Portugaly, Dipankar Ray, Patrice Simard and Ed Snelson:  //**Counterfactual Reasoning and Learning Systems**//, arXiv:1209.2355, September 2012.
  
 \\ \\
 [[http://arxiv.org/abs/1209.2355|arXiv link]] [[http://arxiv.org/abs/1209.2355|arXiv link]]
 +<html>&nbsp;&nbsp;</html>
 +Latest: 
 [[http://leon.bottou.org/publications/djvu/tr-2012-09-12.djvu|tr-2012-09-12.djvu]] [[http://leon.bottou.org/publications/djvu/tr-2012-09-12.djvu|tr-2012-09-12.djvu]]
 [[http://leon.bottou.org/publications/pdf/tr-2012-09-12.pdf|tr-2012-09-12.pdf]] [[http://leon.bottou.org/publications/pdf/tr-2012-09-12.pdf|tr-2012-09-12.pdf]]
 [[http://leon.bottou.org/publications/psgz/tr-2012-09-12.ps.gz|tr-2012-09-12.ps.gz]] [[http://leon.bottou.org/publications/psgz/tr-2012-09-12.ps.gz|tr-2012-09-12.ps.gz]]
 </box> </box>
 +
 +A version of this text was published in [[bottou-jmlr-2013| 2013 (JMLR)]].
  
   @techreport{tr-bottou-2012,   @techreport{tr-bottou-2012,
-    author = {Bottou, L\'eon and Peters, Jonas and Qui\~{n}onero-Candela, Joaquin and  +    author = {Bottou, L\'eon and Peters, Jonas and Qui{\~n}onero-Candela, Joaquin and  
-              Charles, Denis X. and Chickering, D. Max and Portugualy, Elon and +              Charles, Denis X. and Chickering, D. Max and Portugaly, Elon and 
               Ray, Dipankar and Simard, Patrice and Snelson, Ed},               Ray, Dipankar and Simard, Patrice and Snelson, Ed},
     title = {Counterfactual Reasoning and Learning Systems},     title = {Counterfactual Reasoning and Learning Systems},
Line 33: Line 28:
     url = {http://leon.bottou.org/papers/tr-bottou-2012},     url = {http://leon.bottou.org/papers/tr-bottou-2012},
   }   }
 +  
 +
  
papers/tr-bottou-2012.1347451983.txt.gz · Last modified: 2012/09/12 08:13 by leonb

Page Tools