Autocurricula and the Emergence of Innovation from Social Interaction: A Manifesto for Multi-Agent Intelligence Research
This work addresses the problem of fostering innovation in AI systems for researchers in multi-agent intelligence, but it is incremental as it builds on existing evolutionary and social dynamics concepts.
The paper proposes that multi-agent systems can generate their own learning curriculum through social interactions, leading to innovation and increasingly complex challenges.
Evolution has produced a multi-scale mosaic of interacting adaptive units. Innovations arise when perturbations push parts of the system away from stable equilibria into new regimes where previously well-adapted solutions no longer work. Here we explore the hypothesis that multi-agent systems sometimes display intrinsic dynamics arising from competition and cooperation that provide a naturally emergent curriculum, which we term an autocurriculum. The solution of one social task often begets new social tasks, continually generating novel challenges, and thereby promoting innovation. Under certain conditions these challenges may become increasingly complex over time, demanding that agents accumulate ever more innovations.