Perspective Taking in Deep Reinforcement Learning Agents

Frontiers in Computational Neuroscience

Raul Vicente¹  Ardi Tampuu¹  Aqeel Labash¹  Tambet Matiisen¹  Jaan Aru² 

[1] Computational Neuroscience Lab, Institute of Computer Science, University of Tartu, Tartu, Estonia;Institute of Biology, Humboldt University of Berlin, Berlin, Germany;

关键词: deep reinforcement learning;    theory of mind;    perspective taking;    multi-agent;    artificial intelligence;   

DOI  :  10.3389/fncom.2020.00069

来源: DOAJ

【 摘 要 】

Perspective taking is the ability to take into account what the other agent knows. This skill is not unique to humans as it is also displayed by other animals like chimpanzees. It is an essential ability for social interactions, including efficient cooperation, competition, and communication. Here we present our progress toward building artificial agents with such abilities. We implemented a perspective taking task inspired by experiments done with chimpanzees. We show that agents controlled by artificial neural networks can learn via reinforcement learning to pass simple tests that require some aspects of perspective taking capabilities. We studied whether this ability is more readily learned by agents with information encoded in allocentric or egocentric form for both their visual perception and motor actions. We believe that, in the long run, building artificial agents with perspective taking ability can help us develop artificial intelligence that is more human-like and easier to communicate with.

【 授权许可】

Unknown