On the Computability of Solomonoff Induction and Knowledge-Seeking

@article{Leike2015OnTC,
  title={On the Computability of Solomonoff Induction and Knowledge-Seeking},
  author={J. Leike and Marcus Hutter},
  journal={ArXiv},
  year={2015},
  volume={abs/1507.04124}
}
  • J. Leike, Marcus Hutter
  • Published 2015
  • Computer Science, Mathematics
  • ArXiv
  • Solomonoff induction is held as a gold standard for learning, but it is known to be incomputable. We quantify its incomputability by placing various flavors of Solomonoff's prior M in the arithmetical hierarchy. We also derive computability bounds for knowledge-seeking agents, and give a limit-computable weakly asymptotically optimal reinforcement learning agent. 
    11 Citations
    On the computability of Solomonoff induction and AIXI
    • 2
    • PDF
    On the Computability of AIXI
    • 8
    • PDF
    Beyond Almost-Sure Termination
    • 5
    • PDF
    Putnam’s Diagonal Argument and the Impossibility of a Universal Learning Machine
    • 3
    • PDF
    Nonparametric General Reinforcement Learning
    • 12
    • PDF
    Absolutely No Free Lunches!
    • G. Belot
    • Computer Science, Mathematics
    • Theor. Comput. Sci.
    • 2020
    • PDF
    Loss Bounds and Time Complexity for Speed Priors
    • 5
    • PDF

    References

    SHOWING 1-10 OF 23 REFERENCES
    On the Computability of AIXI
    • 8
    • PDF
    Optimality Issues of Universal Greedy Agents with Static Priors
    • 26
    • PDF
    Asymptotic non-learnability of universal agents with computable horizon functions
    • 20
    • PDF
    Universal Knowledge-Seeking Agents for Stochastic Environments
    • 34
    • PDF
    Universal Knowledge-Seeking Agents
    • 21
    • Highly Influential
    • PDF
    Bad Universal Priors and Notions of Optimality
    • 26
    • PDF
    Asymptotically Optimal Agents
    • 31
    • PDF
    A Theory of Universal Artificial Intelligence based on Algorithmic Complexity
    • 61
    • PDF
    On the Relation between Descriptional Complexity and Algorithmic Probability
    • P. Gács
    • Computer Science
    • Theor. Comput. Sci.
    • 1983
    • 54
    • PDF
    (Non-)Equivalence of Universal Priors
    • 23
    • PDF