Videos

Generative Adversarial Nets – Fresh Machine Learning #2



Siraj Raval

This episode of Fresh Machine Learning is all about a relatively new concept called a Generative Adversarial Network. A model continuously tries to fool another model, until it can do so with ease. At that point, it can generate novel, authentic looking data! Very exciting stuff.

The demo code for this video is a set of adversarial Gaussian Distribution Curves in Python using Theano and PyPlot:

https://github.com/llSourcell/Generative-Adversarial-Network-Demo

I created a Slack channel for us, sign up here:
https://wizards.herokuapp.com/

I introduce two papers in this video

Generative Adversarial Networks:

https://arxiv.org/pdf/1406.2661v1.pdf

and the associated code:

https://github.com/goodfeli/adversarial

Generative Adversarial Text-to-Image Synthesis:

https://arxiv.org/pdf/1605.05396v2.pdf

and it’s associated code is here:

https://github.com/reedscot/icml2016

Another really cool repo using GANs:

https://github.com/Newmu/dcgan_code

Great explanation of GANs:

http://soumith.ch/eyescream/

Live demo of a GAN:

http://cs.stanford.edu/people/karpathy/gan/

One more really great description of generative models:

https://openai.com/blog/generative-models/

I love you guys! Thanks for watching my videos, I do it for you. I left my awesome job at Twilio and I’m doing this full time now.

I recently created a Patreon page. If you like my videos, feel free to help support my effort here!:
https://www.patreon.com/user?ty=h&u=3191693

Much more to come so please subscribe, like, and comment.
Follow me:
Twitter: https://twitter.com/sirajraval
Facebook: https://www.facebook.com/sirajology Instagram: https://www.instagram.com/sirajraval/ Instagram: https://www.instagram.com/sirajraval/
Signup for my newsletter for exciting updates in the field of AI:
https://goo.gl/FZzJ5w
Hit the Join button above to sign up to become a member of my channel for access to exclusive content!

Source

Similar Posts

26 thoughts on “Generative Adversarial Nets – Fresh Machine Learning #2
  1. Hi Siraj, Your videos which were LIVE on the Deep Learning Playlist, do not have any option to comments…
    Also, for the GAN style transfer code, i am facing this issue … on line…
    rec_z = inference_network(p_x, latent_dim, n_layer_inf, n_hidden_inf, eps_dim )
    ValueError: Variable inference/Repeat/fully_connected_1/weights already exists, disallowed. Did you mean to set reuse=True in VarScope?

  2. I wonder, how much training data is (was) needed to create those photo realistic images? I'd like to try some GAN work recreationally on different types of data and I am curious about the data I would need

  3. Thanks Siraj for this video! But I have a question about GAN, how does the discriminator improve the results of the generator? I can't understand the link between the output of the discriminator and the input of the generator, how the D is affecting the G. Can you explain this please?

  4. @ 0:30 "models like support vector machines, neural nets and HIDDEN MAKOV CHAIN". I thought they(HMMs) were generative being the sequential versions of navie bayes that they are.

Comments are closed.

WP2Social Auto Publish Powered By : XYZScripts.com