Robot Learning Theory of Mind through Self-Observation: Exploiting the Intentions-Beliefs Synergy

TL;DR

This paper demonstrates that learning to predict others' intentions and actions in robots is significantly improved when simultaneously learning to attribute beliefs, leveraging self-observation in partially observable environments.

Contribution

It introduces a novel approach that combines learning intentions and beliefs, enhancing theory of mind capabilities in robots through self-observation and deep learning.

Findings

Faster and more accurate predictions when beliefs are learned simultaneously.

Improved learning even with agents having different decision processes.

Higher performance when observing beliefs-driven behaviors.

Abstract

In complex environments, where the human sensory system reaches its limits, our behaviour is strongly driven by our beliefs about the state of the world around us. Accessing others' beliefs, intentions, or mental states in general, could thus allow for more effective social interactions in natural contexts. Yet these variables are not directly observable. Theory of Mind (TOM), the ability to attribute to other agents' beliefs, intentions, or mental states in general, is a crucial feature of human social interaction and has become of interest to the robotics community. Recently, new models that are able to learn TOM have been introduced. In this paper, we show the synergy between learning to predict low-level mental states, such as intentions and goals, and attributing high-level ones, such as beliefs. Assuming that learning of beliefs can take place by observing own decision and beliefs estimation processes in partially observable environments and using a simple feed-forward deep learning model, we show that when learning to predict others' intentions and actions, faster and more accurate predictions can be acquired if beliefs attribution is learnt simultaneously with action and intentions prediction. We show that the learning performance improves even when observing agents with a different decision process and is higher when observing beliefs-driven chunks of behaviour. We propose that our architectural approach can be relevant for the design of future adaptive social robots that should be able to autonomously understand and assist human partners in novel natural environments and tasks.