MacKay, David J. C. (1992) Information-based objective functions for active data selection. Neural Computation, 4 (4). pp. 590-604. ISSN 0899-7667 http://resolver.caltech.edu/CaltechAUTHORS:MACnc92c
- Published Version
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:MACnc92c
Learning can be made more efficient if we can actively select particularly salient data points. Within a Bayesian learning framework, objective functions are discussed that measure the expected informativeness of candidate measurements. Three alternative specifications of what we want to gain information about lead to three different criteria for data selection. All these criteria depend on the assumption that the hypothesis space is correct, which may prove to be their main weakness.
|Additional Information:||© 1992 Massachusetts Institute of Technology. Received 17 July 1991; accepted 15 November 1991. Posted Online March 13, 2008. I thank Allen Knutsen, Tom Loredo, Marcus Mitchell, and the referees for helpful feedback. This work was supported by a Caltech Fellowship and a Studentship from SERC, UK.|
|Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Tony Diaz|
|Deposited On:||18 Jun 2009 19:52|
|Last Modified:||26 Dec 2012 10:54|
Repository Staff Only: item control page