Critical behavior in a cross-situational lexicon learning scenario
Tilles, P. F. C., Fontanari, J. F.
–arXiv.org Artificial Intelligence
The problem of early word-learning has been subject of philosophical controversy for centuries [1]. The always visionary Augustine argued that the child makes the connections between words and their referents by understanding the referential intentions of others, thus anticipating the modern theory of mind in about fifteen centuries [2]. In the 17th century, Locke's empiricism supported the associationist viewpoint, which contends that the mechanism of word learning is sensitivity to covariation, i.e., if two events occur at the same time, they become associated. Here we examine a radical offshoot of the associationist approach to lexicon acquisition termed crosssituational or observational learning [3], which asserts that the meaning of a word can be determined by looking for something in common across all observed uses of that word [4]. In other words, learning takes place through the statistical sampling of the contexts in which a word appears.
arXiv.org Artificial Intelligence
Jun-13-2012
- Country:
- Asia > Singapore (0.04)
- North America > United States
- Massachusetts > Middlesex County
- Cambridge (0.04)
- New York (0.04)
- Massachusetts > Middlesex County
- South America > Brazil
- São Paulo (0.05)
- Genre:
- Research Report (0.40)
- Technology: