Sign in

Building robust classifiers through generation of confident out of distribution examples

By Kumar Sricharan and Ashok Srivastava
Deep learning models are known to be overconfident in their predictions on out of distribution inputs. 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.... Show more
December 1, 2018
Loading PDF…
Loading full text...
Similar articles
Loading recommendations...