diff options
| author | Thibaut Horel <thibaut.horel@gmail.com> | 2015-02-06 16:56:08 -0500 |
|---|---|---|
| committer | Thibaut Horel <thibaut.horel@gmail.com> | 2015-02-06 16:56:08 -0500 |
| commit | 8d7b0d99cfadc38a15b0f63d28d0edd024e8c5f0 (patch) | |
| tree | 3e66fa1524852dd03400378e4615c8d57925a5dc /paper/sections/results.tex | |
| parent | 2d9b15a9ec01aa5cf90bf92edf24aab4fe6970c8 (diff) | |
| download | cascades-8d7b0d99cfadc38a15b0f63d28d0edd024e8c5f0.tar.gz | |
Fix corollary constant in approx sparsity
Diffstat (limited to 'paper/sections/results.tex')
| -rw-r--r-- | paper/sections/results.tex | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/paper/sections/results.tex b/paper/sections/results.tex index 1688aae..c83e4d6 100644 --- a/paper/sections/results.tex +++ b/paper/sections/results.tex @@ -265,7 +265,7 @@ As before, edge recovery is a consequence of upper-bounding $\|\theta - \hat \th Under the same assumptions as Theorem~\ref{thm:approx_sparse} the number of measurements verifies: \begin{equation} n > \frac{9}{\alpha\gamma^2\epsilon^2}\left(1+ - \frac{4}{\epsilon^2}\| \theta^* - \theta^*_{\lfloor + \frac{16}{\epsilon^2}\| \theta^* - \theta^*_{\lfloor s\rfloor}\|_1\right)s\log m \end{equation} then similarly: ${\cal S}^*_{\eta + \epsilon} \subset \hat {\cal S}_\eta |
