Computable Variants of AIXI which are More Powerful than AIXItl
This work provides a more flexible and powerful computable alternative to AIXI for researchers in artificial general intelligence, though it is incremental relative to existing variants.
The paper introduces Unlimited Computable AI (UCAI), a family of computable variants of AIXI that addresses limitations of AIXItl by supporting models like typed lambda calculus and removing program length constraints, resulting in a more powerful framework.
This paper presents Unlimited Computable AI, or UCAI, that is a family of computable variants of AIXI. UCAI is more powerful than AIXItl, that is a conventional family of computable variants of AIXI, in the following ways: 1) UCAI supports models of terminating computation, including typed lambda calculus, while AIXItl only supports Turing machine with timeout t, which can be simulated by typed lambda calculus for any t; 2) unlike UCAI, AIXItl limits the program length to l.