Risk-Aware Active Inverse Reinforcement Learning

作者： Scott Niekum , Daniel S. Brown , Yuchen Cui

DOI:

关键词: Work (physics) 、 Inverse reinforcement learning 、 Variety (cybernetics) 、 State space 、 Table (database) 、 Task (computing) 、 Robot 、 Machine learning 、 Artificial intelligence 、 Computer science 、 Active learning (machine learning)

摘要: Active learning from demonstration allows a robot to query a human for specific types of input to achieve efficient learning. Existing work has explored a variety of active query …