@fchollet In your AE Tutorial, I saw that you are training with fixed noise over the inputs. That is not quite right ..
-
-
Replying to @recurseparadox @fchollet
Fresh/Different noise for every presentation of the sample is the standard and theoretically sound method. FWIW.
1 reply 0 retweets 0 likes -
Replying to @recurseparadox
haha no. You're saying "only right way to train a model is w synthetic data". Only possible for the simplest kind of noise.
1 reply 0 retweets 0 likes
Replying to @fchollet @recurseparadox
just because it's a tutorial doesn't mean we should not make it useful in the real world.
8:17 AM - 2 Aug 2016
0 replies
0 retweets
0 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.