Search papers, labs, and topics across Lattice.
The paper critiques the inability of current AI models to achieve autonomous learning, highlighting their reliance on large, curated datasets and lack of embodied interaction. It proposes a novel learning architecture (A, B, M) inspired by cognitive science, integrating observational learning (System A), active learning (System B), and meta-control (System M) for adaptive behavior. The authors suggest that this architecture, drawing inspiration from biological adaptation, could enable AI systems to learn more effectively in dynamic, real-world environments.
Current AI's hunger for curated data may be solved by a new architecture inspired by human cognition that flexibly switches between observation, active behavior, and meta-control.
We critically examine the limitations of current AI models in achieving autonomous learning and propose a learning architecture inspired by human and animal cognition. The proposed framework integrates learning from observation (System A) and learning from active behavior (System B) while flexibly switching between these learning modes as a function of internally generated meta-control signals (System M). We discuss how this could be built by taking inspiration on how organisms adapt to real-world, dynamic environments across evolutionary and developmental timescales.
BAIR
Meta AI