diff options
| author | Jean Pouget-Abadie <jean.pougetabadie@gmail.com> | 2015-03-30 18:14:46 -0400 |
|---|---|---|
| committer | Jean Pouget-Abadie <jean.pougetabadie@gmail.com> | 2015-03-30 18:14:46 -0400 |
| commit | 0b4c21f29459e7947b624b813e24f9fd811219a2 (patch) | |
| tree | 466700361ec3e1f148aeb81f7ea2dd1004e5da63 | |
| parent | 842a1685a5fcf9545cc3ef0e79b4ab67781d9ab9 (diff) | |
| download | cascades-0b4c21f29459e7947b624b813e24f9fd811219a2.tar.gz | |
fixing typo
| -rw-r--r-- | paper/rebuttal.txt | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/paper/rebuttal.txt b/paper/rebuttal.txt index 0911224..8ac4e91 100644 --- a/paper/rebuttal.txt +++ b/paper/rebuttal.txt @@ -25,7 +25,7 @@ constraints: we might need for applications (e.g. simulations). This is application-dependent and somewhat out of the scope of our paper, but it is easy to prove that if we normalize the learned parameters to sum up to - one after solving (2), the l2 guarantee of Theorem 2 looses + one after solving (2), the l2 guarantee of Theorem 2 loses a multiplicative factor at most √s. - If we know from the beginning that we will need the learned parameters to |