Geometric topology: a space of structure on a manifold .

What does it mean for two structures to be nearby?

Consider the space with the smooth weak topology, where and are smooth manifolds with dimension respectively.

Definition. for smooth weak topology. Choose , where is compact, finitely many charts covering . Then, is the set of all such that on each of the charts, are close in , i.e. for all . That is, the partial derivatives of order less than or equal to differ by less than .

Now, let be a closed -manifold. A structure is , such that , , .

So, dev determines

Definition.

Then, notice , so we can give it the subspace topology.

Special case. Suppose is injective. Let . Identify with . Then, dev becomes the inclusion . Then, is nearby in the weak topology if on a compact set , is close the the identity in the topology.

We have a smalll neighborhood in the weak topology if we have a big and a small .

This is a stronger version of the Holonomy Theorem: If is a closed manifold, is open.

Lemma 3.3 Let be a properly convex manifold. Let , the tautological line bundle. Let be a flow function, i.e. there is a flow such that . Then is Hessian convex ( (positive definite matrix)) if and only if is a Hessian convex surface.

Example Consider . This is convex in the old sense, because every secant lies on one side. But, it’s not Hessian convex because .

Definition A smooth hypersurface is Hessian convex if , the tangent plane to at , there exists a neighborhood of , above , where is the graph of and (positive definite).

Now, we give our main result of today:

Openness for Properly Convex Structures (Koszul \~ 1962) Let be a closed -manifold.
Then is open.

Sketch proof: Suppose , so , holonomy . Suppose close to . Let , close to . This implies that nearby .

We can to the same thing for and .

Similarly, there exist nearby dev maps, . Then, notice , where denote the cone over .

So, there exists with lift where is close the the identity on a large compact set in . Thus is close to 0, and is close the the identity.

Define , a level set of the convexity function . Then, is a hypersurface, and is Hessian convex by the lemma.

So, . Since and identity, we use the chain rule to see that is Hessian convex in , since is compact. Then, by the lemma, we know that is the zero set of a convexity flow function on , and from the theorem from last time, we know that is properly convex.

Previous Post: Lecture 15: Tautological Line Bundle


Next Post: Lecture 17: Geometric Transitions