Pandaemonium Architecture 6.0

In this course

Augustus Odena is a researcher known for early work identifying failure modes and instability problems in generative adversarial networks (GANs), particularly the production of local artifacts and texture blobs rather than coherent objects. His 2016 paper is cited here as foundational documentation of GAN pathologies that the course readings use to establish the baseline of problems that subsequent techniques like Spectral Normalization attempt to solve. Odena's work matters to this seminar because it exemplifies how machine learning systems can produce failures that are neither random nor trivial, forcing researchers to develop both analytical and regularization methods to understand and stabilize deep generative models.

Mentioned in 1 reading

Andrew Brock

et al., LARGE SCALE GAN TRAINING FOR HIGH FIDELITY NATURAL IMAGE SYNTHESIS

4

Of particular relevance to our work is Spectral Normalization (Miyato et al., 2018), which enforces Lipschitz continuity on D by normalizing its parameters with running estimates of their first singular values, inducing backwards dynamics t…Much previous work has investigated GAN stability from a variety of analytical angles and on toy problems, but the instabilities we observe occur for settings which are stable at small scale, necessitating direct analysis at large scale. We…We note that some failure modes of our partially-trained models are distinct from those previously observed. Most previous failures involve local artifacts (Odena et al., 2016), images consisting of texture blobs instead of objects (Saliman…

+1 more

Appears alongside

People mentioned in the same passages — sorted by co-occurrence weight.