Help Me Explore: Minimal Social Interventions for Graph-Based Autotelic Agents

TL;DR

This paper introduces a social interaction protocol called Help Me Explore (HME) and a graph-based autotelic agent, GANGSTR, demonstrating that social guidance significantly enhances skill acquisition in manipulation tasks.

Contribution

It proposes a novel social learning framework combining Piagetian and Vygotskian perspectives and develops GANGSTR, a new graph-based agent capable of complex goal decomposition.

Findings

GANGSTR masters complex configurations with few social interventions.

HME improves learning efficiency in manipulation tasks.

Social guidance extends the agent's capabilities beyond individual learning limits.

Abstract

In the quest for autonomous agents learning open-ended repertoires of skills, most works take a Piagetian perspective: learning trajectories are the results of interactions between developmental agents and their physical environment. The Vygotskian perspective, on the other hand, emphasizes the centrality of the socio-cultural environment: higher cognitive functions emerge from transmissions of socio-cultural processes internalized by the agent. This paper argues that both perspectives could be coupled within the learning of autotelic agents to foster their skill acquisition. To this end, we make two contributions: 1) a novel social interaction protocol called Help Me Explore (HME), where autotelic agents can benefit from both individual and socially guided exploration. In social episodes, a social partner suggests goals at the frontier of the learning agent knowledge. In autotelic episodes, agents can either learn to master their own discovered goals or autonomously rehearse failed social goals; 2) GANGSTR, a graph-based autotelic agent for manipulation domains capable of decomposing goals into sequences of intermediate sub-goals. We show that when learning within HME, GANGSTR overcomes its individual learning limits by mastering the most complex configurations (e.g. stacks of 5 blocks) with only few social interventions.