Feature Directed Active Learning
Published in ICAPS Workshop on Explainable AI in Planning (XAIP 2019), 2019
Recommended citation: Gopalakrishnan, S., Soni, U., & Kambhampati, S. (2019). Feature Directed Active Learning. In ICAPS XAIP Workshop. https://arxiv.org/abs/1907.05436
Abstract
When learning domain dynamics from human teachers, naive uncertainty sampling often queries uninformative states. We propose feature-directed active query selection focusing on human-interpretable feature boundaries.
Recommended citation: Gopalakrishnan, S., Soni, U., & Kambhampati, S. (2019). Feature Directed Active Learning. In ICAPS XAIP Workshop.