Position: Artificial Intelligence Needs Meta Intelligence -- the Case for Metacognitive AI

TL;DR

This paper advocates for integrating metacognitive strategies into AI systems to enhance their accuracy, security, and efficiency, supported by a federated learning case study and a new software framework.

Contribution

It introduces the concept of metacognitive AI, discusses design challenges, and provides a practical framework and case study demonstrating improved AI performance.

Findings

Enhanced learning efficiency in federated learning case study

Improved AI security and effectiveness through metacognitive strategies

Development of a software framework for metacognitive AI applications

Abstract

This position paper argues for metacognition as a general design principle for creating more accurate, secure, and efficient AI. The metacognitive solution involves systems monitoring their own states and judiciously allocating resources depending on each problem instance's difficulty or cost of mistakes. Drawing inspiration both from past work on resource-rational AI and from well-documented metacognitive strategies in psychology and cognitive science, we identify specific challenges in embedding these strategies into AI design and highlight open theoretical and implementation problems. We showcase these principles through a tangible example of improved learning efficiency, effectiveness, and security in a Federated Learning (FL) case study. We show how these principles can be translated into practice with a novel software framework developed specifically to allow the community to design, deploy, and experiment with metacognition-enabled AI applications.