[
home |
resume |
contact |
science |
art |
personal |
email
]
Davies, J. & Gagné, J. (2010). Estimating quantitative magnitudes using semantic similarity.
Conference of the American Association for Artificial Intelligence workshop on Visual
Representations and Reasoning (AAAI-10-VRR), 14--19.
Cite this for:
- SYSTEM: Visuo
- When estimating quantitative magnitudes, using a semantically-related
base for analogy works better than using a semantically-unrelated base.
BibTex Entry:
@InProceedings{DaviesGagne2010,
author = {Davies, Jim and Gagné, Jonathan},
title = {Estimating quantitative magnitudes using semantic similarity},
booktitle = {Conference of the American Association for Artificial Intelligence workshop on Visual
Representations and Reasoning (AAAI-10-VRR),
pages = {14--19},
year = {2010},
editor = {}
}
Download: [
PDF ]
Abstract
We present an AI called Visuo that guesses quantitative
visuospatial magnitudes (e.g., heights, lengths) given
adjective-noun pairs as input (e.g., “big hat”). It uses a
database of tagged images as memory and infers unexperienced
magnitudes by analogy with semantically related
concepts in memory. We show that transferring
width-height ratios from a semantically-related concept
yields significantly lower error rates than using dissimilar
concepts when predicting the width-height ratios of novel
inputs.
Back to Jim Davies's list of
publications.
JimDavies
(
jim@jimdavies.org
)