Nuts and Bolts of WGANs, Kantorovich-Rubistein Duality, Earth Movers Distance
If you have seen the last video on GANs, it may have created an illusion that training GANs is super easy. But when you sit back and train them, they can be quite a headache. Stability of Generator is extremely fragile in most of the scenarios. So today, we are going to present a well-written paper by Martin Arjovsky, Soumith Chintala and L´eon Bottou from Facebook AI Research Group.
I am also impressed by blogs written by Vincent Herrmann and Alex Irpan.
Video Covers :
Why do we need Earth Mover's Distance?
How does WGANs or wasserstein gans gets rid of Mode Collapse? (ofcourse clipping weights is brutal)
Why is training WGANs more stable?
What is kantorovich Rubinstien duality?
Cool visualizations to understand difficulties in estimation of a probability distribution.
References
1. https://vincentherrmann.github.io/blog/wasserstein/
2. https://www.alexirpan.com/2017/02/22/wasserstein-gan.html
3. https://lilianweng.github.io/lil-log/2017/08/20/from-GAN-to-WGAN.html
4. Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., & Courville, A. C. (2017). Improved training of wasserstein gans. In Advances in Neural Information Processing Systems (pp. 5767-5777).
5. https://arxiv.org/abs/1701.07875
6. Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A., & Chen, X. (2016). Improved techniques for training gans. In Advances in Neural Information Processing Systems (pp. 2234-2242).
Please consider supporting the channel on Patreon page :
https://www.patreon.com/crazymuse
Thanks to my current patrons for supporting the work :
1. Parth Parikh
2. Laher .D
3. Sean Marrett
Видео Nuts and Bolts of WGANs, Kantorovich-Rubistein Duality, Earth Movers Distance канала Crazymuse
I am also impressed by blogs written by Vincent Herrmann and Alex Irpan.
Video Covers :
Why do we need Earth Mover's Distance?
How does WGANs or wasserstein gans gets rid of Mode Collapse? (ofcourse clipping weights is brutal)
Why is training WGANs more stable?
What is kantorovich Rubinstien duality?
Cool visualizations to understand difficulties in estimation of a probability distribution.
References
1. https://vincentherrmann.github.io/blog/wasserstein/
2. https://www.alexirpan.com/2017/02/22/wasserstein-gan.html
3. https://lilianweng.github.io/lil-log/2017/08/20/from-GAN-to-WGAN.html
4. Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., & Courville, A. C. (2017). Improved training of wasserstein gans. In Advances in Neural Information Processing Systems (pp. 5767-5777).
5. https://arxiv.org/abs/1701.07875
6. Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A., & Chen, X. (2016). Improved techniques for training gans. In Advances in Neural Information Processing Systems (pp. 2234-2242).
Please consider supporting the channel on Patreon page :
https://www.patreon.com/crazymuse
Thanks to my current patrons for supporting the work :
1. Parth Parikh
2. Laher .D
3. Sean Marrett
Видео Nuts and Bolts of WGANs, Kantorovich-Rubistein Duality, Earth Movers Distance канала Crazymuse
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
![WGAN implementation from scratch (with gradient penalty)](https://i.ytimg.com/vi/pG0QZ7OddX4/default.jpg)
![Estimating the Wasserstein Metric - Jonathan Niles-Weed](https://i.ytimg.com/vi/glHy9z19mI0/default.jpg)
![Improved Training of Wasserstein GANs Presentation Group J](https://i.ytimg.com/vi/UcIOkhUGWkw/default.jpg)
![Variational Autoencoders](https://i.ytimg.com/vi/9zKuYvjFFS8/default.jpg)
![Detecting Anomalies Using Statistical Distances | SciPy 2018 | Charles Masson](https://i.ytimg.com/vi/U7xdiGc7IRU/default.jpg)
![Earth Mover's Distance and Maximum Mean Discrepancy | Unsupervised Learning for Big Data](https://i.ytimg.com/vi/Bk84wAkunpo/default.jpg)
![Introduction to the Wasserstein distance](https://i.ytimg.com/vi/CDiol4LG2Ao/default.jpg)
![Divergence and curl: The language of Maxwell's equations, fluid flow, and more](https://i.ytimg.com/vi/rB83DpBJQsE/default.jpg)
![09. Regularized Wasserstein Distances & Minimum Kantorovich Estimators. Marco Cuturi](https://i.ytimg.com/vi/T9O6T5WHdC8/default.jpg)
![The Wasserstein Metric a.k.a Earth Mover's Distance: A Quick and Convenient Introduction](https://i.ytimg.com/vi/ymWDGzpQdls/default.jpg)
![Understand the Math and Theory of GANs in ~ 10 minutes](https://i.ytimg.com/vi/J1aG12dLo4I/default.jpg)
![The incredible inventions of intuitive AI | Maurice Conti](https://i.ytimg.com/vi/aR5N2Jl8k14/default.jpg)
![Talking PyTorch and Careers in AI: Soumith Chintala and Mat Leonard](https://i.ytimg.com/vi/VyI1cRJi0b8/default.jpg)
![3D Hand Pose Estimation Using Convolutional Neural Networks](https://i.ytimg.com/vi/aE7kW4b6CjA/default.jpg)
![AI generated faces - StyleGAN explained!](https://i.ytimg.com/vi/4LNO8nLxF4Y/default.jpg)
![What are GANs (Generative Adversarial Networks)?](https://i.ytimg.com/vi/TpMIssRdhco/default.jpg)
![Deep Learning 35: (2) Wasserstein Generative Adversarial Network (WGAN): Wasserstein metric](https://i.ytimg.com/vi/y8LGAhzCOxQ/default.jpg)
![GAN Lecture 6 (2018): WGAN, EBGAN](https://i.ytimg.com/vi/3JP-xuBJsyc/default.jpg)
![Daniel Kuhn: "Wasserstein Distributionally Robust Optimization: Theory and Applications in Machi..."](https://i.ytimg.com/vi/Hhhtsi2tKlI/default.jpg)
![Intuitively Understanding the KL Divergence](https://i.ytimg.com/vi/SxGYPqCgJWM/default.jpg)