← Search

Ziyin Liu

2 accepted papers

2019

Deep Gamblers: Learning to Abstain with Portfolio Theory

NeurIPS 2019poster

We deal with the selective classification problem (supervised-learning problem with a rejection option), where we want to achieve the best performance at a certain level of coverage of the data. We transform the original $m$-class classification problem to (m+1)-class where the (m+1)-th class repres…