The Recurrent Temporal Restricted Boltzmann Machine is a promising probabilistic model for processing temporal data. It has been shown to learn physical dynamics from videos (e.g. bouncing balls), but its ability to process sequential data has not been tested on symbolic tasks. Here we assess its capabilities on learning sequences of letters corresponding to English words. It emerged that the model is able to extract local transition rules between items of a sequence (i.e. English graphotactic rules), but it does not seem to be suited to encode a whole word.

Assessment of sequential boltzmann machines on a lexical processing task

Testolin A.;Sperduti A.;Stoianov I.;Zorzi M.
2012

Abstract

The Recurrent Temporal Restricted Boltzmann Machine is a promising probabilistic model for processing temporal data. It has been shown to learn physical dynamics from videos (e.g. bouncing balls), but its ability to process sequential data has not been tested on symbolic tasks. Here we assess its capabilities on learning sequences of letters corresponding to English words. It emerged that the model is able to extract local transition rules between items of a sequence (i.e. English graphotactic rules), but it does not seem to be suited to encode a whole word.
2012
ESANN 2012 proceedings, 20th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
20th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2012
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11577/3469776
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? ND
  • OpenAlex ND
social impact