Bottom-Up Hidden Tree Markov Model is a highly expressive model for tree-structured data. Unfortunately, it cannot be used in practice due to the intractable size of its state-transition matrix. We propose a new approximation which lies on the Tucker factorisation of tensors. The probabilistic interpretation of such approximation allows us to define a new probabilistic model for tree-structured data. Hence, we define the new approximated model and we derive its learning algorithm. Then, we empirically assess the effective power of the new model evaluating it on two different tasks. In both cases, our model outperforms the other approximated model known in the literature.

Bayesian Tensor Factorisation for Bottom-up Hidden Tree Markov Models / Castellana D.; Bacciu D.. - ELETTRONICO. - 2019-:(2019), pp. 1-8. (Intervento presentato al convegno 2019 International Joint Conference on Neural Networks, IJCNN 2019 tenutosi a hun nel 2019) [10.1109/IJCNN.2019.8851851].

Bayesian Tensor Factorisation for Bottom-up Hidden Tree Markov Models

Castellana D.
;
2019

Abstract

Bottom-Up Hidden Tree Markov Model is a highly expressive model for tree-structured data. Unfortunately, it cannot be used in practice due to the intractable size of its state-transition matrix. We propose a new approximation which lies on the Tucker factorisation of tensors. The probabilistic interpretation of such approximation allows us to define a new probabilistic model for tree-structured data. Hence, we define the new approximated model and we derive its learning algorithm. Then, we empirically assess the effective power of the new model evaluating it on two different tasks. In both cases, our model outperforms the other approximated model known in the literature.
2019
Proceedings of the International Joint Conference on Neural Networks
2019 International Joint Conference on Neural Networks, IJCNN 2019
hun
2019
Castellana D.; Bacciu D.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

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

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1304184
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 7
  • ???jsp.display-item.citation.isi??? 7
social impact