The dazzling success of neural networks over natural language processing systems is imposing a urgent need to control their behavior with simpler, more direct declarative rules. In this paper, we propose Pat-in-the-loop as a model to control a specific class of syntax-oriented neural networks by adding declarative rules. In Pat-in-the-loop, distributed tree encoders allow to exploit parse trees in neural networks, heat parse trees visualize activation of parse trees, and parse subtrees are used as declarative rules in the neural network. A pilot study on question classification showed that declarative rules representing human knowledge can be effectively used in these neural networks.3

Pat-in-the-loop: Syntax-based neural networks with activation visualization and declarative control / Zanzotto, F. M.; Onorati, D.; Tommasino, P.; Santilli, A.; Ranaldi, L.; Fallucchi, F.. - 2742:(2020), pp. 112-118. (Intervento presentato al convegno 2020 Italian Workshop on Explainable Artificial Intelligence, XAI.it 2020 tenutosi a Online).

Pat-in-the-loop: Syntax-based neural networks with activation visualization and declarative control

Onorati D.;Santilli A.;
2020

Abstract

The dazzling success of neural networks over natural language processing systems is imposing a urgent need to control their behavior with simpler, more direct declarative rules. In this paper, we propose Pat-in-the-loop as a model to control a specific class of syntax-oriented neural networks by adding declarative rules. In Pat-in-the-loop, distributed tree encoders allow to exploit parse trees in neural networks, heat parse trees visualize activation of parse trees, and parse subtrees are used as declarative rules in the neural network. A pilot study on question classification showed that declarative rules representing human knowledge can be effectively used in these neural networks.3
2020
2020 Italian Workshop on Explainable Artificial Intelligence, XAI.it 2020
natural language processing; visualization; explainable artificial intelligence
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
Pat-in-the-loop: Syntax-based neural networks with activation visualization and declarative control / Zanzotto, F. M.; Onorati, D.; Tommasino, P.; Santilli, A.; Ranaldi, L.; Fallucchi, F.. - 2742:(2020), pp. 112-118. (Intervento presentato al convegno 2020 Italian Workshop on Explainable Artificial Intelligence, XAI.it 2020 tenutosi a Online).
File allegati a questo prodotto
File Dimensione Formato  
Zanzotto_Pat_2020.pdf

accesso aperto

Tipologia: Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza: Creative commons
Dimensione 759.41 kB
Formato Adobe PDF
759.41 kB Adobe PDF

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/11573/1643156
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact