Seite - 177 - in Differential Geometrical Theory of Statistics
Bild der Seite - 177 -
Text der Seite - 177 -
Entropy2016,18, 433
Thereforeonehas
gâ(Ï(X),Y)=Xgâ(Ο,Y)âgâ(Ο,D0XY).
Since thequadruple
(M,gâ,D0,Dâ)
isaduallyïŹatpaironehas the identity
gâ(Ï(X),Y)= gâ(DâXΟ,Y).
Thusweget theexpectedconclusion,viz
Ï(X)=DâXΟ.
Conversely letusassumethat thereexistsavectorΟ satisfyingthe identity
Ï(X)=DâXΟ.
That leads to the identity
gâ(DâXΟ,Y)=Xgâ(Ο,Y)âg(Ο,D0XY).
Inotherwordsonehas
gÏâHyp(M,D0).
Thisends theproofofCorollary2.
Thesetofgâ-symmetricgaugetransformation isdenotedbyÎŁ(gâ).
Wehavethecanonical isomorphism
ÎŁ(gâ) Ïâ gÏâRie(M). (12)
NowwedeïŹnethesets
ZË1Ï(Aâ,Aâ)=ÎŁ(gâ)â©Z1Ï(AâAâ),
BË1Ï(Aâ,Aâ)=ÎŁ(gâ)â©B1Ï(Aâ,Aâ).
CombiningLemma3and its corollarywith the isomorphismEquation (12). Thenweobtain
the identiïŹcations
ZË1Ï(Aâ,Aâ)=Hes(M,Dâ),
BË1Ï(Aâ,Aâ)=Hyp(M,Dâ).
Reminder.
Werecall that ahyperbolicmanifold (oraKoszulmanifold) isÎŽKV-exactHessianmanifold (M,g,D).
It is easily seen that the set ofpositivehyperbolic structures ina locallyïŹatmanifold (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,Dâ)
Hyp(M,Dâ)âH 2
KV(Aâ,R),
177
Differential Geometrical Theory of Statistics
- Titel
- Differential Geometrical Theory of Statistics
- Autoren
- Frédéric Barbaresco
- Frank Nielsen
- Herausgeber
- MDPI
- Ort
- Basel
- Datum
- 2017
- Sprache
- englisch
- Lizenz
- CC BY-NC-ND 4.0
- ISBN
- 978-3-03842-425-3
- Abmessungen
- 17.0 x 24.4 cm
- Seiten
- 476
- Schlagwörter
- Entropy, Coding Theory, Maximum entropy, Information geometry, Computational Information Geometry, Hessian Geometry, Divergence Geometry, Information topology, Cohomology, Shape Space, Statistical physics, Thermodynamics
- Kategorien
- Naturwissenschaften Physik