Just wrote 5 pages for my new paper! I finally have something interesting to present! 😺 I show a geometric interpretation of classifiers. WGAN-GP is different from the Wasserstein Distance and I show what it is actually optimizing. I also talk about RaGANs.
6
12
214
Replies
Neat! I just found out why the original Relativistic paired GANs didn't work well except when using a gradient penalty. This paper is going to have lots of interesting tidbits.
2
1
25
I also found gradient penalties which work better than WGAN-GP 🙀 and there is a simple theoretical reason to why this is the case. I will release the paper (with code and blog) on Monday or Tuesday next week! 😸
2
3
49
I think my paper from next week will bring some heated debates...
1
0
19
After some heated debate👩🚒🔥🚒, it was decided that the paper will be send for AISTATS2020, but we will only send it to Arxiv on October 14. So it's delayed by a week.😿
2
1
17
Ever wondered how to make sense of gradient penalized or 1-Lipschitz discriminators (using spectral normalization) in GANs other than Wasserstein's? My next-week paper will provide an answer and show a framework to derive all these GANs directly.
1
2
13
My new paper is finally coming out tomorrow!!!🙀
The title is:
"Connections between Support Vector Machines, Wasserstein distance and gradient-penalty GANs"😸
Everything is connected!🤯
2
9
53
@gokstudio
I haven't checked conferences yet. The simple enough kind that even I who isn't great at geometry can understand. It's a pretty neat result.
0
0
1
