Design for a Darwinian Brain: Part 2. Cognitive Architecture
This addresses the challenge of cumulative adaptation in reinforcement learning and developmental robotics, though it appears incremental as it builds on existing concepts like intrinsic motivation.
The paper tackles the problem of enabling open-ended lifetime learning in AI by proposing a cognitive architecture that can specify an unlimited range of behaviors and stochastically explore adjacent possible ones, with partial implementation on a humanoid robot.
The accumulation of adaptations in an open-ended manner during lifetime learning is a holy grail in reinforcement learning, intrinsic motivation, artificial curiosity, and developmental robotics. We present a specification for a cognitive architecture that is capable of specifying an unlimited range of behaviors. We then give examples of how it can stochastically explore an interesting space of adjacent possible behaviors. There are two main novelties; the first is a proper definition of the fitness of self-generated games such that interesting games are expected to evolve. The second is a modular and evolvable behavior language that has systematicity, productivity, and compositionality, i.e. it is a physical symbol system. A part of the architecture has already been implemented on a humanoid robot.