Skip to main content
10 events
when toggle format what by license comment
Jan 28, 2017 at 22:56 vote accept Terry Tao
Jan 28, 2017 at 19:01 comment added Robert Bryant @TerryTao: Yes, they are the left-invariant forms of the affine group. You can think of this group as embedded in $\mathrm{GL}(5,\mathbb{R})$ as the matrix $$g=\begin{pmatrix}1&0&0&0&0\\ x&e_1&e_2&e_3&e_4\end{pmatrix}=\begin{pmatrix}1&0\\x&e\end{pmatrix}.$$ Then the equation $$\mathrm{d}g = g\begin{pmatrix}0&0\\ \omega&\theta\end{pmatrix} = g\,\gamma$$ is the matrix form of the first structure equation, ie., $\gamma = g^{-1}\,\mathrm{d}g$. The second equation is then just $$\mathrm{d}\gamma=-\gamma\wedge\gamma.$$
Jan 28, 2017 at 18:26 comment added Terry Tao Thanks for this! I am slowly going through the calculations. Is there a Lie group interpretation of the 1-forms $\omega^i$ and $\theta^j_i$? It looks like they should somehow be associated to the affine group ${\bf R}^4 \rtimes GL(4,{\bf R})$, but I don't see the precise relation yet.
Jan 28, 2017 at 11:50 history edited Robert Bryant CC BY-SA 3.0
Put in a construction of polynomial solutions and corrected some typos
Jan 27, 2017 at 21:42 history edited Robert Bryant CC BY-SA 3.0
Fixed some incorrect formulae
Jan 27, 2017 at 15:17 comment added Deane Yang Very nice application of exterior differential systems.
Jan 27, 2017 at 14:31 history edited Robert Bryant CC BY-SA 3.0
added 8451 characters in body
Jan 27, 2017 at 14:23 comment added Robert Bryant @TerryTao: Ok, I've put it in. If you have questions, or something needs clarification, please let me know.
Jan 27, 2017 at 3:22 comment added Terry Tao I would be interested in seeing more details, thanks. Right now I don't see how the bundle $F$ is interacting with the subspaces $W$.
Jan 27, 2017 at 2:18 history answered Robert Bryant CC BY-SA 3.0