Combining dimensions and features in similarity-based representations
Keyword(s):
Data Set
◽
This paper develops a new representational model of similarity data that combines continuous dimensions with discrete features. An algorithm capable of learning these representations is described, and a Bayesian model selection approach for choosing the appropriate number of dimensions and features is developed. The approach is demonstrated on a classic data set that considers the similarities between the numbers 0 through 9.
2020 ◽
Vol 90
(14)
◽
pp. 2631-2654
2018 ◽
Vol 78
(5)
◽
pp. 3413-3423
◽
2012 ◽
Vol 28
(3)
◽
pp. 919-931
◽
2004 ◽
Vol 5
(4)
◽
pp. 379-387
◽
2017 ◽
Vol 33
(1)
◽
pp. 235-248
◽
2009 ◽
Vol 3
(1)
◽
pp. 39-61
◽
2017 ◽
Vol 67
(1)
◽
pp. 3-23
◽