Referential uncertainty and word learning in high-dimensional, continuous meaning spaces

Abstract

Abstract-This paper discusses lexicon word learning in highdimensional meaning spaces from the viewpoint of referential uncertainty. We investigate various state-of-the-art Machine Learning algorithms and discuss the impact of scaling, representation and meaning space structure. We demonstrate that current Machine Learning techniques successfully deal with high-dimensional meaning spaces. In particular, we show that exponentially increasing dimensions linearly impact learner performance and that referential uncertainty from word sensitivity has no impact.

Learning to Parse Grounded Language using Reservoir Computing

“…Grounded language learning has often focused on the acquisition of categories and concepts [15], for example in the spatial language [16] and color domains [17]. In that work, centroid-based classifiers with distance (Voronoi tessellation) or other Machine Learning algorithms (neural networks, Bayesian models etc.)…”

Computational models of tutor feedback in language acquisition

“…KNN is the only tested algorithm for continuous meaning cross-situational learning [19]. Centroid algorithms are the goto method for interactive learning [14], [15], [20].…”

“…In interactive learning researchers often propose to use prototypes for the learner's lexicon [14], [15], [20] and to update these prototypes iteratively as new samples of word object mappings are collected by the learner. a) Representation: The learner stores prototypes p ∈ R 3 .…”