Page - 184 - in Differential Geometrical Theory of Statistics
Image of the Page - 184 -
Text of the Page - 184 -
Entropy2016,18, 433
Lemma5. The integern(D0)doesnotdependonthe choice of g∈Rie(M).
AnIdea.
Wefixametric tensor g. For everyg∗ ∈Rie(M) there is auniqueg-symmetricvectorbundlemorphism
φ∈Σ(g) such that
g∗(X,Y)= g(φ(X),Y).
Therefore,wehave
φ−1◦M(D0,Dg)=M(g∗,D0,Dg∗),
φ−1M(g,D0,Dg)=M(g∗,D0,Dg∗).
Nowonedefines thenumerical invariantn(M).
Definition34.
n(M)=max{n(D)|D∈SLC(M)} .
Given a Koszul connection∇ the vector space of∇-parallel differential 2-forms is denoted
byΩ∇2 (M).
Everydualpair (M,g,∇,∇∗)givesrise to the linear isomorphisms
(1) : M(∇,∇∗)
M(g,∇,∇∗) [ψ]→ qψ∈S ∇
2 (M),
(2) : M(g,∇,∇∗) ψ→ωψ∈Ω∇2 (M).
The isomorphism(1)derives fromthe linearmap
(1∗) : ψ→ qψ(X,Y)= 12[g(ψ(X),Y)+g(X,ψ(Y))].
The isomorphism(2) isdefinedby
(2∗) : ψ→ωψ(X,Y)= 12[g(ψ(X),Y)−g(X,ψ(Y))].
Proposition6. Let (M,g,∇,∇∗)beadualpair. The inclusionmap
M(g,∇,∇∗)→M(∇,∇∗)
induced the split short exact sequence
(∗∗∗∗∗) : 0→Ω∇2 (M)→M(∇,∇∗)→S∇2 (M)→0.
Reminder.
According our previous notation elements ofΩ∇2 (M) are∇-geodesic symplectic foliations. Those of
S∇2 (M)are∇-geodesicRiemannian foliations. Thusweapplymethodsof the informationgeometry to relate the
gaugegeometryandthedifferential topolgy.
Digressions.
Ourconstructionmayopen tonewdevelopments.Hereare someunexploredperspectives.
(a) A∇-geodesic symplectic foliationω∈Ω∇might carryricher structures suchasKahlerianstructures.
184
Differential Geometrical Theory of Statistics
- Title
- Differential Geometrical Theory of Statistics
- Authors
- Frédéric 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