Transcription of Wasserstein Generative Adversarial Networks
{{id}} {{{paragraph}}}
Example: confidence
Wasserstein Generative Adversarial Networks
Wasserstein Generative Adversarial NetworksMartin Arjovsky1 Soumith Chintala2L eon Bottou1 2 AbstractWe introduce a new algorithm named WGAN,an alternative to traditional GAN training. Inthis new model, we show that we can improvethe stability of learning, get rid of problems likemode collapse, and provide meaningful learningcurves useful for debugging and hyperparametersearches. Furthermore, we show that the cor-responding optimization problem is sound, andprovide extensive theoretical work highlightingthe deep connections to different distances be-tween IntroductionThe problem this paper is concerned with is that of unsu-pervised learning. Mainly, what does it mean to learn aprobability distribution? The classical answer to this is tolearn a probability density. This is often done by defininga parametric family of densities(P ) Rdand finding theone that maximized the likelihood on our data: if we havereal data examples{x(i)}mi=1, we would solve the problemmax Rd1mm i=1logP (x(i))If the real data distributionPradmits a density andP is thedistribution of the parametrized densityP , then, asymp-totically, this amounts to minimizing the Kullback-LeiblerdivergenceKL(Pr P ).
a distance or divergence ˆ(P ;P r). The most fundamen-tal difference between such distances is their impact on the convergence of sequences of probability distributions. A sequence of distributions (P t) t2N converges if and only if there is a distribution P 1such that ˆ(P t;P 1)tends to zero, something that depends on how exactly the ...
Loading..
Tags:
Information
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document: