Emergent constraints on word-learning: a computational perspective

@article{Regier2003EmergentCO,
  title={Emergent constraints on word-learning: a computational perspective},
  author={Terry Regier},
  journal={Trends in Cognitive Sciences},
  year={2003},
  volume={7},
  pages={263-268}
}
  • T. Regier
  • Published 2003
  • Psychology, Medicine
  • Trends in Cognitive Sciences
In learning the meanings of words, children are guided by a set of constraints that give privilege to some potential meanings over others. These word-learning constraints are sometimes viewed as part of a specifically linguistic endowment. However, several recent computational models suggest concretely how word-learning - constraints included - might emerge from more general aspects of cognition, such as associative learning, attention and rational inference. This article reviews these models… Expand
Learning Words and Their Meanings from Unsegmented Child-directed Speech
TLDR
A new Bayesian segmentation model is presented that incorporates aspects of word learning and is compared to a model that ignores word meanings, suggesting that the non-linguistic context may supply important information for learning word segmentations as well as word meanings. Expand
The Emergence of Words: Attentional Learning in Form and Meaning
TLDR
An associative exemplar-based model is presented that accounts for the improvement at word learning without a change in mechanism, and explains these improvements in terms of increased attention to relevant aspects of form and meaning, which reduces memory interference. Expand
Learning Words in Space and Time
TLDR
Across three experiments, the mechanistic basis of the suspicious-coincidence effect showed that the presentation of multiple subordinate-level exemplars led to narrow generalization only when the exemplars were presented simultaneously, even when the number of exemplars was increased from three to six. Expand
A unified model of early word learning: Integrating statistical and social cues
TLDR
It is argued that statistical and social cues can be seamlessly integrated to facilitate early word learning and a unified model is presented that is able to make use of different kinds of embodied social cues in the statistical learning framework. Expand
The emergence of links between lexical acquisition and object categorization: a computational study
  • Chen Yu
  • Computer Science
  • Connect. Sci.
  • 2005
TLDR
This work proposes and implements a computational model of how word learning influences the formation of object categories to which those words refer, and shows promising improvements in both word-to-world mapping and perceptual categorization. Expand
Word learning emerges from the interaction of online referent selection and slow associative learning.
TLDR
An alternative in which referent selection is an online process and independent of long-term learning is presented, which suggests that association learning buttressed by dynamic competition can account for much of the literature and suggests more sophisticated ways of describing the interaction between situation- and developmental-time processes. Expand
A Statistical Associative Account of Vocabulary Growth in Early Word Learning
TLDR
This work suggests that using lexical knowledge accumulated in prior statistical learning could play an important role in vocabulary growth, and is the first model that attempts to simulate the effects of cumulative knowledge on subsequent learning using realistic data collected from child-caregiver interactions. Expand
The Role of Embodied Intention in Early Lexical Acquisition
TLDR
This work is the first model of word learning that not only learns lexical items from raw multisensory signals to closely resemble infant language development from natural environments, but also explores the computational role of social cognitive skills in lexical acquisition. Expand
USAGE-BASED AND FORM-FOCUSED LANGUAGE ACQUISITION: The associative learning of constructions, learned attention, and the limited L2 endstate
TLDR
This section concludes by illustrating the combined operation of these factors in first and second language acquisition of English grammatical morphemes, a particular illustration of a broader claim that they control the acquisition of all linguistic constructions. Expand
Sensitivity to sampling in Bayesian word learning.
TLDR
A new study testing the proposal that word learning may be best explained as an approximate form of Bayesian inference finds that this result follows naturally from a Bayesian analysis, but not from other statistical approaches such as associative word-learning models. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 57 REFERENCES
How children learn the meanings of words
How do children learn that the word "dog" refers not to all four-legged animals, and not just to Ralph, but to all members of a particular species? How do they learn the meanings of verbs likeExpand
Evidence against a dedicated system for word learning in children
TLDR
The findings show that fast mapping is not limited to word learning, suggesting that the capacity to learn and retain new words is the result of learning and memory abilities that are not specific to language. Expand
Learning Nouns and Adjectives: A Connectionist Account
TLDR
An alternative account of why do children learn nouns such as cup faster than dimensional adjectives such as big is examined, one which relies instead on properties of the semantic categories to be learned and of the word-learning task itself. Expand
Becoming a word learner : a debate on lexical acquisition
TLDR
This book presents an Emergentist, Coalition Model for Word Learning: Mapping Words to Objects is Product of the Internation of Multiple Cues and Training the Problem Space in Early Word Learning. Expand
Rapid Word Learning in 13- and 18-Month-Olds.
A number of theorists have argued that the productive naming explosion results from advances in abilities that underlie language learning (e.g., the realization that words are symbols, changes inExpand
Modelling Parsing Constraints with High-dimensional Context Space
TLDR
It is proposed that HAL's high-dim ensional context space can be used to provide a basic categorisation of semantic and grammatical concepts, model certain aspects of morphological ambiguity in verbs, and provide an account of semantic context effects in syntactic processing. Expand
The child as word learner
concepts of reference point and polarity, which presuppose concepts of dimension of comparison and zero point. These concepts can be probed nonlinguistically. For example, animals can be taught toExpand
A computational study of cross-situational techniques for learning word-to-meaning mappings
TLDR
This paper presents a computational study of part of the lexical-acquisition task faced by children, namely the acquisition of word-to-meaning mappings, and presents an implemented algorithm for solving this problem, illustrating its operation on a small example. Expand
Acquiring the Mapping from Meaning to Sounds
TLDR
This work models the process of children's early vocabulary development using a recurrent neural network trained to map a set of plan vectors to associated sequences of phonemes, representing the phonological structure of the surface forms, and evaluates the role of the similarity structures of the target forms and the input forms on the evolution of the network's vocabulary. Expand
Beyond fast mapping: young children's extensions of novel words and novel facts.
TLDR
Two studies are reported in which 2-4-year-old children learned novel words and novel facts for unfamiliar objects and then were asked to extend the words and facts to additional exemplars of the training objects and results show that by 2 years of age, children honor the necessary extendibility of novel count nouns but are uncertain about the extendability of arbitrary facts. Expand
...
1
2
3
4
5
...