KDnuggets Home » News » 2015 » n23 ( < n22 | n24 > )

KDnuggets™ News 15:n23, Jul 22: Deep Learning Adversarial Examples Myths & Facts; Stop Hiring Data Scientists Until …

Deep Learning Adversarial Examples - facts and myths; Stop Hiring Data Scientists Until; arXiv.org and the 24 Hour Research Cycle; Statistics Denial Myth and Repackaging Statistics.

Features |  Tutorials |  Opinions |  Interviews |  News |  Webcasts |  Courses |  Meetings |  Jobs |  Publications |  Tweets | Quote
 Tutorials, Overviews, How-Tos
 Webcasts and Webinars
 Top Tweets

Myth: Deep learning is more vulnerable to adversarial examples than other kind of machine learning. Fact: So far we have been able to generate adversarial examples for every model we have tested. Ian Goodfellow (Google), Deep Learning Adversarial Examples - Clarifying Misconceptions