An Experiment in Temporal Language Learning

Gerasymova, Kateryna; Spranger, Michael

doi:10.1007/978-1-4614-3064-3_12

Cited by 6 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tellier and Abend et al [51,52] present a Bayesian approach to learning categorial grammars from utterances that were annotated with a semantic representation expressed in first-order logic. Gerasymova & Spranger [53,54] investigate the acquisition of holophrastic constructions, item-based constructions and abstract constructions for Russian aspectual marking in a tutor-learner language game setting [55,56]. Holophrastic constructions are learned by a straightforward mapping operation between an observed form and its meaning.…”

Section: Mechanistic Models Of Usage-based Language Acquisitionmentioning

confidence: 99%

“…In general, the existing mechanistic models that learn operational computational construction grammars (i.e. class one and two) have explored interesting ideas on a rather small scale, either because they were limited to specific linguistic phenomena [33,53,54,[57][58][59][60]65,66], or because of the limited morpho-syntactic and semantic complexity of the input utterances [50,52,61,62,64,69,[71][72][73]. In all of the aforementioned work except [67,68], either a segmentation of the input utterances, a lexicon or a set of predefined grammatical categories was provided.…”

Section: Mechanistic Models Of Usage-based Language Acquisitionmentioning

confidence: 99%

See 1 more Smart Citation

Modelling constructivist language acquisition through syntactico-semantic pattern finding

Doumen,

Beuls,

Van Eecke

2024

R. Soc. Open Sci.

View full text Add to dashboard Cite

The constructivist acquisition of language by children has been elaborately documented by researchers in psycholinguistics and cognitive science. However, despite the centrality of human-like communication in the field of artificial intelligence, no faithful computational operationalizations of the mechanisms through which children learn language exist to date. In this article, we fill part of this void by introducing a mechanistic model of the constructivist acquisition of language through syntactico-semantic pattern finding. Concretely, we present a methodology for learning grammars based on similarities and differences in the form and meaning of linguistic observations alone. The resulting grammars consist of form-meaning mappings of variable extent and degree of abstraction, called constructions, which facilitate both language comprehension and production. Applying our methodology to the CLEVR benchmark dataset, we provide a proof of concept that demonstrates the online, incremental, data-efficient, transparent and effective learning of item-based construction grammars from utterance–meaning pairs.

show abstract

Section: Mechanistic Models Of Usage-based Language Acquisitionmentioning

confidence: 99%

Section: Mechanistic Models Of Usage-based Language Acquisitionmentioning

confidence: 99%

Modelling constructivist language acquisition through syntactico-semantic pattern finding

Doumen,

Beuls,

Van Eecke

2024

R. Soc. Open Sci.

View full text Add to dashboard Cite

show abstract

“…One research strand has explored how to learn symbolic grammars in the framework of language games. Models have been developed in particular for spatial language [6], [18], temporal language learning [19] and events [20]. A lot of this research focuses on framing language learning as learning constructions; it is often motivated by insights from Cognitive Linguistics, and in particular Cognitive Semantics, Construction Grammar and Usage-based linguistics.…”

Section: Related Workmentioning

confidence: 99%

Learning to Parse Grounded Language using Reservoir Computing

Hinaut

Spranger

2019

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)

Self Cite

View full text Add to dashboard Cite

Recently new models for language processing and learning using Reservoir Computing have been popular. However, these models are typically not grounded in sensorimotor systems and robots. In this paper, we develop a model of Reservoir Computing called Reservoir Parser (ResPars) for learning to parse Natural Language from grounded data coming from humanoid robots. Previous work showed that ResPars is able to do syntactic generalization over different sentences (surface structure) with the same meaning (deep structure). We argue that such ability is key to guide linguistic generalization in a grounded architecture. We show that ResPars is able to generalize on grounded compositional semantics by combining it with Incremental Recruitment Language (IRL). Additionally, we show that ResPars is able to learn to generalize on the same sentences, but not processed word by word, but as an unsegmented sequence of phonemes. This ability enables the architecture to not rely only on the words recognized by a speech recognizer, but to process the sub-word level directly. We additionally test the model's robustness to word error recognition.

show abstract

“…Our group has done various experiments concerning the acquisition of words and concepts [7], [8] and the acquisition of grammar [9]. In these experiments there is either one robot acting as tutor and another one as learner or a human plays the role of tutor.…”

Section: Introductionmentioning

confidence: 99%

Discovering communication through ontogenetic ritualisation

Spranger

Steels

2014

4th International Conference on Development and Learning and on Epigenetic Robotics

Self Cite

View full text Add to dashboard Cite

The entry into symbolic communication through language, gesture or visual signs is one of the key moments in the mental and social development of infants. It is the point from which they start to have a much better social interaction with their parents, other children and adults, and can begin to observe the massive achievements of cultural accumulation. The question addressed in this paper is how developing robots could autonomously make this important transition in their mental development. Based on observations of the way human infants bootstrap into symbolic communication, we propose that gestural symbolic communication comes before auditory symbolic communication and is discovered through a process of ontogenetic ritualisation. The paper identifies the nature of ontogenetic ritualisation and reports on first experiments to achieve this form of learning in humanoid robots.

show abstract

An Experiment in Temporal Language Learning

Cited by 6 publications

References 8 publications

Modelling constructivist language acquisition through syntactico-semantic pattern finding

Modelling constructivist language acquisition through syntactico-semantic pattern finding

Learning to Parse Grounded Language using Reservoir Computing

Discovering communication through ontogenetic ritualisation

Contact Info

Product

Resources

About