TechRxiv
1/1
2 files

Computable Artificial General Intelligence

preprint
posted on 30.11.2022, 05:47 authored by Michael Timothy BennettMichael Timothy Bennett

Artificial general intelligence (AGI) may herald our extinction, according to AI safety research. Yet claims regarding AGI must rely upon mathematical formalisms -- theoretical agents we may analyse or attempt to build. AIXI appears to be the only such formalism supported by proof that its behaviour is optimal, a consequence of its use of compression as a proxy for intelligence. Unfortunately, AIXI is incomputable and claims regarding its behaviour highly subjective. We argue that this is because AIXI formalises cognition as taking place in isolation from the environment in which goals are pursued (Cartesian dualism). We propose an alternative, supported by proof and experiment, which overcomes these problems. Integrating research from cognitive science with AI, we formalise an enactive model of learning and reasoning to address the problem of subjectivity. This allows us to formulate a different proxy for intelligence, called weakness, which addresses the problem of incomputability. We prove optimal behaviour is attained when weakness is maximised. This proof is supplemented by experimental results comparing weakness and description length (the closest analogue to compression possible without reintroducing subjectivity). Weakness outperforms description length, suggesting it is a better proxy. Furthermore we show that, if cognition is enactive, then minimisation of description length is neither necessary nor sufficient to attain optimal performance. These results undermine the notion that compression is closely related to intelligence. We conclude with a discussion of limitations, implications and future research. There remain several open questions regarding the implementation of scale-able general intelligence. In the short term, these results may be best utilised to improve the performance of existing systems. For example, our results explain why Deepmind's Apperception Engine is able to generalise effectively, and how to replicate that performance by maximising weakness. Likewise in the context of neural networks, our results suggest both limitations of ``scale is all you need", and how those limitations can be overcome.

History

Email Address of Submitting Author

michael.bennett@anu.edu.au

ORCID of Submitting Author

0000-0001-6895-8782

Submitting Author's Institution

Australian National University

Submitting Author's Country

Australia