A Common Framework for Natural Gradient and Taylor based Optimisation using Manifold Theory
classification
💻 cs.LG
stat.ML
keywords
frameworkgradientmethodmethodsnaturaloptimisationtaylorapproximation
read the original abstract
This technical report constructs a theoretical framework to relate standard Taylor approximation based optimisation methods with Natural Gradient (NG), a method which is Fisher efficient with probabilistic models. Such a framework will be shown to also provide mathematical justification to combine higher order methods with the method of NG.
This paper has not been read by Pith yet.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.