Sequential predictions based on algorithmic complexity
Download (357.29 kB)
-
Altmetric Citations
Description
This paper studies sequence prediction based on the monotone Kolmogorov complexity Km = − log m, i.e. based on universal deterministic/one-part MDL. m is extremely close to Solomonoff’s universal prior M, the latter being an excellent predictor in deterministic as well as probabilistic environments, where performance is measured in terms of convergence of posteriors or losses. Despite this closeness to M, it is difficult to assess the prediction quality of m, since little is known about the...[Show more]
dc.contributor.author | Hutter, Marcus | |
---|---|---|
dc.date.accessioned | 2015-08-31T02:27:17Z | |
dc.date.available | 2015-08-31T02:27:17Z | |
dc.identifier.issn | 0022-0000 | |
dc.identifier.uri | http://hdl.handle.net/1885/15035 | |
dc.description.abstract | This paper studies sequence prediction based on the monotone Kolmogorov complexity Km = − log m, i.e. based on universal deterministic/one-part MDL. m is extremely close to Solomonoff’s universal prior M, the latter being an excellent predictor in deterministic as well as probabilistic environments, where performance is measured in terms of convergence of posteriors or losses. Despite this closeness to M, it is difficult to assess the prediction quality of m, since little is known about the closeness of their posteriors, which are the important quantities for prediction. We show that for deterministic computable environments, the “posterior” and losses of m converge, but rapid convergence could only be shown on-sequence; the off-sequence convergence can be slow. In probabilistic environments, neither the posterior nor the losses converge, in general. | |
dc.publisher | Elsevier | |
dc.rights | © 2005 Elsevier Inc. http://www.sherpa.ac.uk/romeo/issn/0022-0000/..."Author's post-print on open access repository after an embargo period of between 12 months and 48 months" from SHERPA/RoMEO site (as at 31/08/15). | |
dc.source | Journal of Computer and System Sciences | |
dc.title | Sequential predictions based on algorithmic complexity | |
dc.type | Journal article | |
local.identifier.citationvolume | 72 | |
dc.date.issued | 2006-02 | |
local.publisher.url | http://www.elsevier.com/ | |
local.type.status | Accepted Version | |
local.contributor.affiliation | Hutter, M., Research School of Computer Science, The Australian National University | |
local.bibliographicCitation.issue | 1 | |
local.bibliographicCitation.startpage | 95 | |
local.bibliographicCitation.lastpage | 117 | |
local.identifier.doi | 10.1016/j.jcss.2005.07.001 | |
Collections | ANU Research Publications |
Download
File | Description | Size | Format | Image |
---|---|---|---|---|
Hutter et al Sequential Predictions based on Algorithmic Complexity 2006.pdf | 357.29 kB | Adobe PDF |
Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.
Updated: 17 November 2022/ Responsible Officer: University Librarian/ Page Contact: Library Systems & Web Coordinator