GANs

Generative Adversarial Networks
www.cs.wisc.edu/~page/cs760/
Goals for the lecture
you should understand the following concepts

• Nash equilibrium
• Minimax game
• Generative adversarial network
Prisoners’ Dilemma
Two suspects in a major crime are held in separate cells. There is enough
evidence to convict each of them of a minor offense, but not enough evidence
to convict either of them of the major crime unless one of them acts as an
informer against the other (defects). If they both stay quiet, each will be
convicted of the minor offense and spend one year in prison. If one and only
one of them defects, she will be freed and used as a witness against the other,
who will spend four years in prison. If they both defect, each will spend three
years in prison.
Players: The two suspects.
Actions: Each player’s set of actions is {Quiet, Defect}.
Preferences: Suspect 1’s ordering of the action profiles, from best to worst, is
(Defect, Quiet) (he defects and suspect 2 remains quiet, so he is freed),
(Quiet, Quiet) (he gets one year in prison), (Defect, Defect) (he gets three
years in prison), (Quiet, Defect) (he gets four years in prison). Suspect 2’s
ordering is (Quiet, Defect), (Quiet, Quiet), (Defect, Defect), (Defect, Quiet).
3 represents best outcome, 0 worst, etc.
Nash Equilibrium
Thanks, Wikipedia.
Another Example
Thanks, Prof. Osborne of U. Toronto, Economics

…And Another Example
Thanks again, Prof. Osborne of U. Toronto, Economics

Minimax with Simultaneous Moves
• maximin value: largest value player can be assured of
without knowing other player’s actions
• minimax value: smallest value other players can force

this player to receive without knowing this player’s action
• minimax is an upper bound on maximin

Key Result
• Utility: numeric reward for actions
• Game: 2 or more players take turns or take simultaneous

actions. Moves lead to states, states have utilities.
• Game is like an optimization problem, but each player

tries to maximize own objective function (utility function)
• Zero-sum game: each player’s gain or loss in utility is

exactly balanced by others’
• In zero-sum game, Minimax solution is same as Nash

Equilibrium
Generative Adversarial Networks
• Approach: Set up zero-sum game between deep nets to
– Generator: Generate data that looks like training set
– Discriminator: Distinguish between real and
synthetic data
• Motivation:
– Building accurate generative models is hard (e.g.,
learning and sampling from Markov net or Bayes net)
– Want to use all our great progress on supervised
learners to do this unsupervised learning task better
– Deep nets may be our favorite supervised learner,
especially for image data, if nets are convolutional
(use tricks of sliding windows with parameter tying,
cross-entropy transfer function, batch normalization)
Does It Work?
Thanks, Ian Goodfellow, NIPS 2016 Tutorial on GANS, for this and most of
what follows…
A Bit More on GAN Algorithm
The Rest of the Details
• Use deep convolutional neural networks for Discriminator
D and Generator G
• Let x denote trainset and z denote random, uniform input
• Set up zero-sum game by giving D the following

objective, and G the negation of it:
• Let D and G compute their gradients simultaneously,

each make one step in direction of the gradient, and
repeat until neither can make progress… Minimax
Not So Fast
• While preceding version is theoretically elegant, in
practice the gradient for G vanishes before we reach
best practical solution
• While no longer true Minimax, use same objective for D

but change objective for G to:
• Sometimes better if instead of using one minibatch at a

time to compute gradient and do batch normalization, we
also have a fixed subset of training set, and use
combination of fixed subset and current minibatch
Comments on GANs
• Poten-ally can use our high-‐powered supervised learners to build
be=er, faster data generators (can they replace MCMC, etc.?)
• While some nice theory based on Nash Equilibria, be=er results in
prac-ce if we move a bit away from the theory
• In general, many in ML community have strong concern that we
don’t really understand why deep learning works, including GANs
• S-ll much research into ﬁguring out why this works be=er than
other genera-ve approaches for some types of data, how we can
improve performance further, how to take these from image data
to other data types where CNNs might not be the most natural
deep network structure

GANs

Uploaded by

Copyright:

Available Formats

GANs

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

GANs

Uploaded by

Copyright:

Available Formats

Generative Adversarial Networks

you should understand the following concepts

Players: The two suspects.

Actions: Each player’s set of actions is {Quiet, Defect}.

Thanks, Prof. Osborne of U. Toronto, Economics

Thanks again, Prof. Osborne of U. Toronto, Economics

• minimax value: smallest value other players can force

• minimax is an upper bound on maximin

• Game: 2 or more players take turns or take simultaneous

• Game is like an optimization problem, but each player

• Zero-sum game: each player’s gain or loss in utility is

• In zero-sum game, Minimax solution is same as Nash

• Let x denote trainset and z denote random, uniform input

• Set up zero-sum game by giving D the following

• Let D and G compute their gradients simultaneously,

• While no longer true Minimax, use same objective for D

• Sometimes better if instead of using one minibatch at a

You might also like