Michael Milford

What Would You Do? Acting by Learning to Predict

2017 · arXiv (Cornell University)

We propose to learn tasks directly from visual demonstrations by learning to predict the outcome of human and robot actions on an environment. We enable a robot to physically perform a human demonstrated task without …

Class Anchor Clustering: A Loss for Distance-based Open Set Recognition

2021

In open set recognition, deep neural networks encounter object classes that were unknown during training. Existing open set classifiers distinguish between known and unknown classes by measuring distance in a network's logit space, assuming that …