Timeline for Can we define exterior derivatives using pushforwards and connections?
Current License: CC BY-SA 3.0
6 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Oct 7, 2016 at 9:36 | comment | added | user40276 | @Sebastian Sorry. I was thinking that you were referring to $d^{\nabla}$ in your answer. This is why I said curvature. | |
Oct 7, 2016 at 9:30 | comment | added | Sebastian | @user40276: You can play a similar game with a connection on a vector bundle to define what is called the exterior derivative $d^\nabla$ for forms with values in that bundle. What you get in this different situation is $(d^\nabla)^2$ is the curvature. | |
Oct 7, 2016 at 9:27 | comment | added | Sebastian | The things is this: using an affine connection gives you a derivative $d_\nabla$ mapping k-forms to k+1-forms. Also it satisfies the Leibniz rule for functions and forms. The important difference is that in general $d_\nabla^2f\neq0,$ as in the remark of Ivan. | |
Oct 7, 2016 at 6:30 | comment | added | user40276 | I think you mean curvature. In this case, it's possible to correct it and get a representation up to homotopy by using a complex of vector bundles. | |
Oct 7, 2016 at 5:18 | comment | added | dorebell | What goes wrong when there's torsion? | |
Oct 7, 2016 at 4:59 | history | answered | Sebastian | CC BY-SA 3.0 |