Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation

Elio Tuci; Tomassino Ferrauto; Gianluca Massera; Stefano Nolfi

doi:10.1007/978-3-642-15193-4_49

Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation

Elio Tuci, Tomassino Ferrauto, Gianluca Massera, Stefano Nolfi

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

2 Citations (Scopus)

Abstract

Population of simulated agents controlled by dynamical neural networks are trained by artificial evolution to access linguistic instructions and to execute them by indicating, touching or moving specific target objects. During training the agent experiences only a subset of all object/action pairs. During post-evaluation, some of the successful agents proved to be able to access and execute also linguistic instructions not experienced during training. This is owe to the development of a semantic space, grounded on the sensory motor capability of the agent and organised in a systematised way in order to facilitate linguistic compositionality and behavioural generalisation.

Original language	English
Title of host publication	From Animals to Animats 11
Subtitle of host publication	11th International Conference on Simulation of Adaptive Behavior, SAB 2010, Paris - Clos Lucé, France, August 25-28, 2010. Proceedings
Pages	523-532
Number of pages	10
Volume	6226
ISBN (Electronic)	978-3-642-15193-4
DOIs	https://doi.org/10.1007/978-3-642-15193-4_49
Publication status	Published - 2010
Event	11th International Conference on Simulation of Adaptive Behavior, SAB 2010 - Paris, France Duration: 25 Aug 2010 → 28 Aug 2010

Conference

Conference	11th International Conference on Simulation of Adaptive Behavior, SAB 2010
Country/Territory	France
City	Paris
Period	25 Aug 2010 → 28 Aug 2010

Access to Document

10.1007/978-3-642-15193-4_49

Permanent link

Persistent link

Cite this

Tuci, E., Ferrauto, T., Massera, G., & Nolfi, S. (2010). Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation. In From Animals to Animats 11: 11th International Conference on Simulation of Adaptive Behavior, SAB 2010, Paris - Clos Lucé, France, August 25-28, 2010. Proceedings (Vol. 6226, pp. 523-532) https://doi.org/10.1007/978-3-642-15193-4_49

@inproceedings{b857f2ebacbe40a68211c485b1b98e7a,

title = "Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation",

abstract = "Population of simulated agents controlled by dynamical neural networks are trained by artificial evolution to access linguistic instructions and to execute them by indicating, touching or moving specific target objects. During training the agent experiences only a subset of all object/action pairs. During post-evaluation, some of the successful agents proved to be able to access and execute also linguistic instructions not experienced during training. This is owe to the development of a semantic space, grounded on the sensory motor capability of the agent and organised in a systematised way in order to facilitate linguistic compositionality and behavioural generalisation.",

author = "Elio Tuci and Tomassino Ferrauto and Gianluca Massera and Stefano Nolfi",

note = "Tuci E., Ferrauto T., Massera G., Nolfi S. Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation. Proceedings of the 11th Int. Conf. on Simulation of Adaptive Behavior (SAB2010), Doncieux S., Girard B., Guillot A., Hallam J., Meyer J-A., Mouret J-B., (Eds), LNCS, Springer Heidelberg, pp. 523-532, 2010.; 11th International Conference on Simulation of Adaptive Behavior, SAB 2010 ; Conference date: 25-08-2010 Through 28-08-2010",

year = "2010",

doi = "10.1007/978-3-642-15193-4_49",

language = "English",

isbn = "978-3-642-15192-7",

volume = "6226",

pages = "523--532",

booktitle = "From Animals to Animats 11",

}

Tuci, E, Ferrauto, T, Massera, G & Nolfi, S 2010, Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation. in From Animals to Animats 11: 11th International Conference on Simulation of Adaptive Behavior, SAB 2010, Paris - Clos Lucé, France, August 25-28, 2010. Proceedings. vol. 6226, pp. 523-532, 11th International Conference on Simulation of Adaptive Behavior, SAB 2010, Paris, France, 25 Aug 2010. https://doi.org/10.1007/978-3-642-15193-4_49

Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation. / Tuci, Elio; Ferrauto, Tomassino; Massera, Gianluca et al.
From Animals to Animats 11: 11th International Conference on Simulation of Adaptive Behavior, SAB 2010, Paris - Clos Lucé, France, August 25-28, 2010. Proceedings. Vol. 6226 2010. p. 523-532.

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

TY - GEN

T1 - Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation

AU - Tuci, Elio

AU - Ferrauto, Tomassino

AU - Massera, Gianluca

AU - Nolfi, Stefano

N1 - Tuci E., Ferrauto T., Massera G., Nolfi S. Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation. Proceedings of the 11th Int. Conf. on Simulation of Adaptive Behavior (SAB2010), Doncieux S., Girard B., Guillot A., Hallam J., Meyer J-A., Mouret J-B., (Eds), LNCS, Springer Heidelberg, pp. 523-532, 2010.

PY - 2010

Y1 - 2010

N2 - Population of simulated agents controlled by dynamical neural networks are trained by artificial evolution to access linguistic instructions and to execute them by indicating, touching or moving specific target objects. During training the agent experiences only a subset of all object/action pairs. During post-evaluation, some of the successful agents proved to be able to access and execute also linguistic instructions not experienced during training. This is owe to the development of a semantic space, grounded on the sensory motor capability of the agent and organised in a systematised way in order to facilitate linguistic compositionality and behavioural generalisation.

AB - Population of simulated agents controlled by dynamical neural networks are trained by artificial evolution to access linguistic instructions and to execute them by indicating, touching or moving specific target objects. During training the agent experiences only a subset of all object/action pairs. During post-evaluation, some of the successful agents proved to be able to access and execute also linguistic instructions not experienced during training. This is owe to the development of a semantic space, grounded on the sensory motor capability of the agent and organised in a systematised way in order to facilitate linguistic compositionality and behavioural generalisation.

U2 - 10.1007/978-3-642-15193-4_49

DO - 10.1007/978-3-642-15193-4_49

M3 - Conference Proceeding (Non-Journal item)

SN - 978-3-642-15192-7

VL - 6226

SP - 523

EP - 532

BT - From Animals to Animats 11

T2 - 11th International Conference on Simulation of Adaptive Behavior, SAB 2010

Y2 - 25 August 2010 through 28 August 2010

ER -

Tuci E, Ferrauto T, Massera G, Nolfi S. Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation. In From Animals to Animats 11: 11th International Conference on Simulation of Adaptive Behavior, SAB 2010, Paris - Clos Lucé, France, August 25-28, 2010. Proceedings. Vol. 6226. 2010. p. 523-532 doi: 10.1007/978-3-642-15193-4_49

Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation

Abstract

Conference

Access to Document

Permanent link

Fingerprint

Cite this