Stacked Structure Learning for Lifted Relational Neural Networks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F18%3A00320432" target="_blank" >RIV/68407700:21230/18:00320432 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1007/978-3-319-78090-0_10" target="_blank" >http://dx.doi.org/10.1007/978-3-319-78090-0_10</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-319-78090-0_10" target="_blank" >10.1007/978-3-319-78090-0_10</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Stacked Structure Learning for Lifted Relational Neural Networks

Popis výsledku v původním jazyce

Lifted Relational Neural Networks (LRNNs) describe relational domains using weighted first-order rules which act as templates for constructing feed-forward neural networks. While previous work has shown that using LRNNs can lead to state-of-the-art results in various ILP tasks, these results depended on hand-crafted rules. In this paper, we extend the framework of LRNNs with structure learning, thus enabling a fully automated learning process. Similarly to many ILP methods, our structure learning algorithm proceeds in an iterative fashion by top-down searching through the hypothesis space of all possible Horn clauses, considering the predicates that occur in the training examples as well as invented soft concepts entailed by the best weighted rules found so far. In the experiments, we demonstrate the ability to automatically induce useful hierarchical soft concepts leading to deep LRNNs with a competitive predictive power.

Název v anglickém jazyce

Stacked Structure Learning for Lifted Relational Neural Networks

Popis výsledku anglicky

Lifted Relational Neural Networks (LRNNs) describe relational domains using weighted first-order rules which act as templates for constructing feed-forward neural networks. While previous work has shown that using LRNNs can lead to state-of-the-art results in various ILP tasks, these results depended on hand-crafted rules. In this paper, we extend the framework of LRNNs with structure learning, thus enabling a fully automated learning process. Similarly to many ILP methods, our structure learning algorithm proceeds in an iterative fashion by top-down searching through the hypothesis space of all possible Horn clauses, considering the predicates that occur in the training examples as well as invented soft concepts entailed by the best weighted rules found so far. In the experiments, we demonstrate the ability to automatically induce useful hierarchical soft concepts leading to deep LRNNs with a competitive predictive power.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GA17-26999S" target="_blank" >GA17-26999S: Hluboké relační učení</a><br>

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Inductive Logic Programming 2017

ISBN

978-3-319-78089-4

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

12

Strana od-do

140-151

Název nakladatele

Springer International Publishing

Místo vydání

Cham

Místo konání akce

Orléans

Datum konání akce

4. 9. 2017

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—