What Can You Do with a Rock? Affordance Extraction via Word Embeddings

Nancy Fulda, Daniel Ricks, Ben Murdoch, David Wingate
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
Autonomous agents must often detect affordances: the set of behaviors enabled by a situation. Affordance extraction is particularly helpful in domains with large action spaces, allowing the agent to prune its search space by avoiding futile behaviors. This paper presents a method for affordance extraction via word embeddings trained on a tagged Wikipedia corpus. The resulting word vectors are treated as a common knowledge database which can be queried using linear algebra. We apply this method
more » ... apply this method to a reinforcement learning agent in a text-only environment and show that affordance-based action selection improves performance in most cases. Our method increases the computational complexity of each learning step but significantly reduces the total number of steps needed. In addition, the agent's action selections begin to resemble those a human would choose.
doi:10.24963/ijcai.2017/144 dblp:conf/ijcai/FuldaRMW17 fatcat:4g7hqbyngvgwvd6erkb66o3pe4