I have an implementation of Bayesian multi-dimensional scaling (MDS), which suffers from the problem (similar to Unique solution for probabilistic PCA ) that there are infinitely number of equally probable coordinates for the items, due to the fact that MDS solutions are invariant to translation, rot…

Breaking symmetries in Bayesian MDS

gBokiau January 23, 2020, 11:23pm 5

Or rather, start with the first, whilst tracking some statistics that are likely sufficient candidates for rendering the solution unique.

Topic		Replies	Views
Unique solution for probabilistic PCA Questions	47	4275	March 8, 2019
Factor Analysis Questions	22	2738	September 1, 2017
Excessive memory - Multiple regression Questions	9	2931	October 24, 2017
Mixture models and breaking class symmetry Questions	11	3932	October 12, 2018
Constrain one variable to be greater than another Questions	14	3519	January 30, 2018