Page - 177 - in Differential Geometrical Theory of Statistics
Image of the Page - 177 -
Text of the Page - 177 -
Entropy2016,18, 433
Thereforeonehas
gā(Ļ(X),Y)=Xgā(ξ,Y)āgā(ξ,D0XY).
Since thequadruple
(M,gā,D0,Dā)
isaduallyļ¬atpaironehas the identity
gā(Ļ(X),Y)= gā(DāXξ,Y).
Thusweget theexpectedconclusion,viz
Ļ(X)=DāXξ.
Conversely letusassumethat thereexistsavectorξ satisfyingthe identity
Ļ(X)=DāXξ.
That leads to the identity
gā(DāXξ,Y)=Xgā(ξ,Y)āg(ξ,D0XY).
Inotherwordsonehas
gĻāHyp(M,D0).
Thisends theproofofCorollary2.
Thesetofgā-symmetricgaugetransformation isdenotedbyĪ£(gā).
Wehavethecanonical isomorphism
Ī£(gā) Ļā gĻāRie(M). (12)
Nowwedeļ¬nethesets
ZĖ1Ļ(Aā,Aā)=Ī£(gā)ā©Z1Ļ(AāAā),
BĖ1Ļ(Aā,Aā)=Ī£(gā)ā©B1Ļ(Aā,Aā).
CombiningLemma3and its corollarywith the isomorphismEquation (12). Thenweobtain
the identiļ¬cations
ZĖ1Ļ(Aā,Aā)=Hes(M,Dā),
BĖ1Ļ(Aā,Aā)=Hyp(M,Dā).
Reminder.
Werecall that ahyperbolicmanifold (oraKoszulmanifold) isΓKV-exactHessianmanifold (M,g,D).
It is easily seen that the set ofpositivehyperbolic structures ina locallyļ¬atmanifold (M,D) is a convex
subset ofHes(M,D).
So show theKoszul geometry is a vanishing theorem in the theory of KVhomology of KValgebroids.
The theoryofhomological statisticalmodel (to be introduced inPartB) is another impact on the information
geometryof theKVcohomology.
At thepresent stepwehave the relations
Hes(M,Dā)
Hyp(M,Dā)āH 2
KV(Aā,R),
177
Differential Geometrical Theory of Statistics
- Title
- Differential Geometrical Theory of Statistics
- Authors
- FrƩdƩric Barbaresco
- Frank Nielsen
- Editor
- MDPI
- Location
- Basel
- Date
- 2017
- Language
- English
- License
- CC BY-NC-ND 4.0
- ISBN
- 978-3-03842-425-3
- Size
- 17.0 x 24.4 cm
- Pages
- 476
- Keywords
- Entropy, Coding Theory, Maximum entropy, Information geometry, Computational Information Geometry, Hessian Geometry, Divergence Geometry, Information topology, Cohomology, Shape Space, Statistical physics, Thermodynamics
- Categories
- Naturwissenschaften Physik