Proving Pushforward Product Isomorphism: M1 x ... x Mk to M1 + ... + Mk

  • Context: Graduate 
  • Thread starter Thread starter robforsub
  • Start date Start date
  • Tags Tags
    Product
Click For Summary

Discussion Overview

The discussion revolves around proving that the pushforward product isomorphism from the tangent space of the product of smooth manifolds M1 x ... x Mk to the direct sum of the tangent spaces T_p1(M1) ⊕ ... ⊕ T_pk(Mk) is indeed an isomorphism. The participants explore various approaches to demonstrate this property, focusing on the linearity and bijectiveness of the proposed map.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant suggests proving the isomorphism by showing that the map a is a bijection, noting that it is already established as a linear map.
  • Another participant proposes taking a natural basis in the image and "guessing" the pre-image to verify the isomorphism.
  • A different participant raises a concern about the necessity of considering the component functions of arbitrary X in the proof.
  • One participant references the manifold structure on the product of two manifolds and discusses the basis associated with charts of the product manifold, suggesting this could aid in the exploration of the proof.

Areas of Agreement / Disagreement

Participants appear to have differing approaches to the proof, with no consensus on the best method to demonstrate the isomorphism. The discussion remains unresolved as various strategies are proposed without agreement on a definitive solution.

Contextual Notes

Participants have not fully addressed the implications of component functions in their arguments, and there may be missing assumptions regarding the nature of the manifolds and the projections involved.

robforsub
Messages
16
Reaction score
0
Let M1,..,Mk be smooth manifolds, and let Pi_j be the projection from M1XM2X...XMk->Mj. Show that the map a:T_(p1,...,pk)(M1XM2X...Mk)->T_p1(M1)\oplus...\oplusT_pk(Mk)

a(X)=(Pi_1*X,Pi_2*X,...,Pi_k*X) is an isomorphism.

The way I am thinking to prove the statement is to show that a is a bijection, since a is already a linear map. And I have no clue how to show it, help needed!
 
Physics news on Phys.org
I would take a natural basis in the image and "guess" what would be its pre-image - then check.
 
But, don't we need to take component functions of arbitrary X into account?
 
Recall the manifold structure on M x N. If (U,f) is a chart of M^m around p and (V,g) is a chart of N^n around q, then (U x V, f x g) is a chart of M x N around (p,q). If ∂/∂x^i is the basis of T_pM associated with the chart (U,f) and ∂/∂y^i is the basis of T_qN associated with the chart (V,g), call (∂/∂x'^1,..., ∂/∂x'^m, ∂/∂y'^1,..., ∂/∂y'^n) the basis associated with the chart (U x V, f x g).

Just doing this should give you some room for your experiments.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 0 ·
Replies
0
Views
2K
  • · Replies 16 ·
Replies
16
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
1
Views
3K
  • · Replies 1 ·
Replies
1
Views
15K
  • · Replies 137 ·
5
Replies
137
Views
21K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 100 ·
4
Replies
100
Views
13K
  • · Replies 39 ·
2
Replies
39
Views
13K