Interactive Learning of State Representation through Natural Language Instruction and Explanation
This addresses the closed-world assumption in robot learning, which is a problem for robots operating in dynamic environments.
The paper tackles the problem of robots lacking a complete world model for new tasks by enabling them to learn state representations through natural language instruction and explanation from humans.
One significant simplification in most previous work on robot learning is the closed-world assumption where the robot is assumed to know ahead of time a complete set of predicates describing the state of the physical world. However, robots are not likely to have a complete model of the world especially when learning a new task. To address this problem, this extended abstract gives a brief introduction to our on-going work that aims to enable the robot to acquire new state representations through language communication with humans.