Integrating Spatial Concepts into a Probabilistic Concept Web

Date

2015-07-31

Author

Celikkanat, Hande
Şahin, Erol
Kalkan, Sinan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

180
views

0
downloads

In this paper, we study the learning and representation of grounded spatial concepts in a probabilistic concept web that connects them with other noun, adjective, and verb concepts. Specifically, we focus on the prepositional spatial concepts, such as "on", "below", "left", "right", "in front of" and "behind". In our prior work (Celikkanat et al., 2015), inspired from the distributed highly-connected conceptual representation in human brain, we proposed using Markov Random Field for modeling a concept web on a humanoid robot. For adequately expressing the unidirectional (i.e., non-symmetric) nature of the spatial propositions, in this work, we propose a extension of the Markov Random Field into a simple hybrid Markov Random Field model, allowing both undirected and directed connections between concepts. We demonstrate that our humanoid robot, iCub, is able to (i) extract meaningful spatial concepts in addition to noun, adjective and verb concepts from a scene using the proposed model, (ii) correct wrong initial predictions using the connectedness of the concept web, and (iii) respond correctly to queries involving spatial concepts, such as ball-left-of-the-cup.

Subject Keywords

Concepts, Concept web, Spatial concepts, Prepositions, Markov random field

URI

https://hdl.handle.net/11511/43124

DOI

https://doi.org/10.1109/icar.2015.7251465

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Verb concepts from affordances Kalkan, Sinan; Yuerueten, Onur; Borghi, Anna M.; Şahin, Erol (John Benjamins Publishing Company, 2014-01-01) In this paper, we investigate how the interactions of a robot with its environment can be used to create concepts that are typically represented by verbs in language. Towards this end, we utilize the notion of affordances to argue that verbs typically refer to the generation of a specific type of effect rather than a specific type of action. Then, we show how a robot can form these concepts through interactions with the environment and how humans can use these concepts to ease their communication with the r...
Formation of adjective, noun and verb concepts through affordances Yürüten, Onur; Kalkan, Sinan; Şahin, Erol; Department of Computer Engineering (2012) In this thesis, we study the development of linguistic concepts (corresponding to a subset of nouns, verbs and adjectives) on a humanoid robot. To accomplish this goal, we use affordances, a notion first proposed by J.J. Gibson to describe the action possibilities offered to an agent by the environment. Using the affordances formalization framework of Sahin et al., we have implemented a learning system on a humanoid robot and obtained the required data from the sensorimotor experiences of the robot. The sys...
A Probabilistic Concept Web on a Humanoid Robot Çelikkanat, Hande; Orhan, Guner; Kalkan, Sinan (2015-06-01) It is now widely accepted that concepts and conceptualization are key elements towards achieving cognition on a humanoid robot. An important problem on this path is the grounded representation of individual concepts and the relationships between them. In this article, we propose a probabilistic method based on Markov Random Fields to model a concept web on a humanoid robot where individual concepts and the relations between them are captured. In this web, each individual concept is represented using a proto...
The learning of adjectives and nouns from affordance and appearance features Yürüten, Onur; Şahin, Erol; Kalkan, Sinan (SAGE Publications, 2013-8-22) We study how a robot can link concepts represented by adjectives and nouns in language with its own sensorimotor interactions. Specifically, an iCub humanoid robot interacts with a group of objects using a repertoire of manipulation behaviors. The objects are labeled using a set of adjectives and nouns. The effects induced on the objects are labeled as affordances, and classifiers are learned to predict the affordances from the appearance of an object. We evaluate three different models for learning adjecti...
The influence of early monolingual dictionaries and word lists on the standardisation of English spelling Sönmez, Margaret Jeanne M. (Walter de Gruyter GmbH, 2001-01-01) This paper aims to ascertain as precisely as possible what direct influence the early monolingual word lists and dictionaries may have had on the standardisation of English spelling. It is based upon research in which a quantitative analysis of the spellings of lemmas from the nine earliest published such lists and dictionaries was made and followed by a comparison of these lemmas and their spellings with the spellings from a manuscript and printed text of appropriate date. These word lists and dictionaries...

Citation Formats

H. Celikkanat, E. Şahin, and S. Kalkan, “Integrating Spatial Concepts into a Probabilistic Concept Web,” 2015, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/43124.