Open-Ended Semantics Co-Evolving With Spatial Language

Spranger, Michael; Pauw, Simon; Loetzsch, Martin

doi:10.1142/9789814295222_0038

Cited by 13 publications

(16 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We employ applicability functions over space, indexed by spatial relations, to induce probability distributions over possible groundings of a symbolic representation. Spranger et al use a similar approach of selecting relations via competition weighted by applicability functions [17], [18] 1 .…”

Section: Inferring Intended Referentsmentioning

confidence: 99%

“…Spranger et al address learning [17] or co-evoloving [18] language about spatial prepositions, using very similar assumptions to our own about acquiring language, such as the idea that speakers choose semantics that not only apply to an object, but contrast it with other objects. However they use a Fluid Construction Grammar that appears to have mappings between semantic elements and linguistic marker locations predefined rather than learned (although marker morphemes are learned or generated).…”

Section: Inferring Intended Referentsmentioning

confidence: 99%

See 1 more Smart Citation

A generative probabilistic framework for learning spatial language

Dawson

Wright

Rebguns

et al. 2013

2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)

View full text Add to dashboard Cite

The language of space and spatial relations is a rich source of abstract semantic structure. We develop a probabilistic model that learns to understand utterances that describe spatial configurations of objects in a tabletop scene by seeking the meaning that best explains the sentence chosen. The inference problem is simplified by assuming that sentences express symbolic representations of (latent) semantic relations between referents and landmarks in space, and that given these symbolic representations, utterances and physical locations are conditionally independent. As such, the inference problem factors into a symbolgrounding component (linking propositions to physical locations) and a symbol-translation component (linking propositions to parse trees). We evaluate the model by eliciting production and comprehension data from human English speakers and find that our system recovers the referent of spatial utterances at a level of proficiency approaching human performance. I. INTRODUCTIONImagine that a friend asks you to "Bring me the thing toward the far corner of the table." This simple request requires fairly sophisticated cognitive processing. You must first identify that she is referring to something on a table, in particular, one with corners. Then, you must orient the table to distinguish "far" vs. "near" corners. Finally, if there is more than one object near a "far" corner, but one is also very near the edge, you might tend to favor the other one, reasoning that it would have been easy to ask for "the one near the edge".We model a version of this problem, with a particular focus on recovering the referent of spatial utterances like "the thing toward the far corner of the table", where the only information available about the intended object is its position relative to a landmark in the scene. Beginning with no knowledge about the meanings of words, but equipped with a small vocabulary of spatial relations (e.g., containment, proximity, ordering in cardinal directions) and abstract representations of objects and their parts (e.g., a table can be represented as a line with ends and a middle, or as a rectangle with corners, quadrants and edges), our model learns probabilistic correspondences between sentences and abstract spatial relations between referents and landmarks by "observing" a teacher repeatedly generating an utterance and pointing to a location in space.Clearly a method that involved supervised learning based on observed propositional semantics would not be developmentally plausible, as children do not get to observe symbolic meaning directly, and so crucially, the abstract relations are never made overt to our learner. Rather, the locations are probabilistically assigned to abstract landmark-relation pairs

show abstract

Section: Inferring Intended Referentsmentioning

confidence: 99%

Section: Inferring Intended Referentsmentioning

confidence: 99%

A generative probabilistic framework for learning spatial language

Dawson

Wright

Rebguns

et al. 2013

2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)

View full text Add to dashboard Cite

show abstract

“…This chapter provides a case study on how a perceptually grounded spatial vocabulary may emerge in a population of autonomous agents, building further on earlier and ongoing work on the emergence of spatial language using language game experiments (Steels, 1995;Steels & Loetzsch, 2009;Spranger et al, 2010;Spranger, 2011a). The methodology applied is similar to earlier chapters of this volume (Steels, 2012).…”

Section: Introductionmentioning

confidence: 99%

The co-evolution of basic spatial terms and categories

Spranger¹

2012

Advances in Interaction Studies

Self Cite

View full text Add to dashboard Cite

This chapter studies how basic spatial categories such as left-right, frontback, far-near or north-south can emerge in a population of robotic agents in co-evolution with terms that express these categories. It introduces various language strategies and tests them first in reconstructions of German spatial terms, then in acquisition experiments to demonstrate the adequacy of the strategy for learning these terms, and finally in language formation experiments showing how a spatial vocabulary and the concepts expressed by it can emerge in a population of embodied agents from scratch.

show abstract

“…Another big topic is how grammatical constructions can be learned. FCG has been used with inductive statistical machine learning algorithms operating over corpora, although most of the work on FCG grammar learning so far has focused on using FCG to learn grammars in situated embodied interactions [Spranger et. al.…”

Section: Discussionmentioning

confidence: 99%

Basics of Fluid Construction Grammar

Steels

2017

View full text Add to dashboard Cite

Fluid Construction Grammar (FCG) is a fully operational computational platform for developing grammars from a constructional perspective. It contains mechanisms for representing grammars and for using them in computational experiments and applications in language understanding, production and learning. FCG can be used by grammar writers who want to test whether their grammar fragments are complete and coherent for the domain they are investigating (for example verb phrases) or who are working in a team and have to share grammar fragments with others. It can be used by computational linguists implementing practical language processing systems or exploring how machine learning algorithms can acquire grammars. This paper introduces some of the basic mechanisms of FCG, illustrated with examples.

show abstract

Open-Ended Semantics Co-Evolving With Spatial Language

Cited by 13 publications

References 9 publications

A generative probabilistic framework for learning spatial language

A generative probabilistic framework for learning spatial language

The co-evolution of basic spatial terms and categories

Basics of Fluid Construction Grammar

Contact Info

Product

Resources

About