State-Reification Networks: Improving Generalization by Modeling the Distribution of Hidden Representations
international conference on machine learning, pp. 3622-3631, 2019.
Machine learning promises methods that generalize well from finite labeled data. However, the brittleness of existing neural net approaches is revealed by notable failures, such as the existence of adversarial examples that are misclassified despite being nearly identical to a training example, or the inability of recurrent sequence-pro...More
PPT (Upload PPT)