Probabilistic Rule Learning through Integer Linear Programming

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F11%3A00178877" target="_blank" >RIV/68407700:21230/11:00178877 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Probabilistic Rule Learning through Integer Linear Programming
Original language description
Recent interest in probabilistic logic as a formalism for machine learning has motivated the formulation of "probabilistic rule learning", where the task is to induce a set of logical rules from a probabilistic database. We start by defining rule learning within propositional, robabilistic logic. Then we show that this problem can be viewed as a regression task with integer variables. This problem is translated into Mixed Linear Programming, which is experimentally shown to provide a speedup over the current implementation. The advantage of this approach is the ability to switch between logically interpretable rule learning with binary coefficients and classical regression with rational coefficients.
Czech name
—
Czech description
—

Project
<a href="/en/project/GAP103%2F10%2F1875" target="_blank" >GAP103/10/1875: Learning from Theories</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2011
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Sborník příspěvků 10. ročníku konference ZNALOSTI 2011
ISBN
978-80-248-2369-0
ISSN
—
e-ISSN
—
Number of pages
9
Pages from-to
45-53
Publisher name
VŠB-TUO
Place of publication
Ostrava
Event location
Stará Lesná
Event date
Jan 31, 2011
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)