Building robust classifiers through generation of confident out of distribution examples

no code implementations • 1 Dec 2018 • Kumar Sricharan, Ashok Srivastava

There have been several pieces of work to address this issue, including a number of approaches for building Bayesian neural networks, as well as closely related work on detection of out of distribution samples.

Improving robustness of classifiers by training against live traffic

no code implementations • 1 Dec 2018 • Kumar Sricharan, Kumar Kallurupalli, Ashok Srivastava

In this paper, we make the following observation: in practice, the out of distribution samples are contained in the traffic that hits a deployed classifier.