Ultimate Cognition à la Gödel

  title={Ultimate Cognition {\`a} la G{\"o}del},
  author={J{\"u}rgen Schmidhuber},
  journal={Cognitive Computation},
  • J. Schmidhuber
  • Published 5 March 2009
  • Computer Science
  • Cognitive Computation
Abstract“All life is problem solving,” said Popper. To deal with arbitrary problems in arbitrary environments, an ultimate cognitive agent should use its limited hardware in the “best” and “most efficient” possible way. Can we formally nail down this informal statement, and derive a mathematically rigorous blueprint of ultimate cognition? Yes, we can, using Kurt Gödel’s celebrated self-reference trick of 1931 in a new way. Gödel exhibited the limits of mathematics and computation by creating a… 
Real-World Limits to Algorithmic Intelligence
It is argued that the world can only be known through the application of rules to observations, and that beyond these rules no knowledge can be obtained about the origin of the authors' observations.
Practical artificial commonsense
The development of a framework for constructing commonsense reasoning systems within constrained time and resources, from present day technology, named Comirit, integrates simulation, logical deduction and machine learning techniques into a coherent whole and is an open-ended framework that allows the integration of any number of additional mechanisms.
Universal Memory Architectures for Autonomous Machines
We propose a self-organizing memory architecture for perceptual experience, capable of supporting autonomous learning and goal-directed problem solving in the absence of any prior information about
Measuring Intelligence through Games
An extension of universal intelligence to deal with night time time, and the use of sampling of the space of games expressed in a suitably biased game description language is argued.
Ethical Artificial Intelligence
How future AI will differ from current AI, the politics of AI, and the ultimate use of AI to help understand the nature of the universe and the authors' place in it are discussed.
PowerPlay: Training an Increasingly General Problem Solver by Continually Searching for the Simplest Still Unsolvable Problem
This work focuses on automatically inventing or discovering problems in a way inspired by the playful behavior of animals and humans, to train a more and more general problem solver from scratch in an unsupervised fashion.
Problems of Self-reference in Self-improving Space-Time Embedded Intelligence
It is shown that in one particular model based on formal logic, naive approaches either lead to incorrect reasoning that allows an agent to put off an important task forever (the procrastination paradox), or fail to allow the agent to justify even obviously safe rewrites (the Lobian obstacle).
Five Answers on Randomness
Five brief and highly biased answers to five questions on rand omness posed by Hector Zenil: Why were you initially drawn to the study of c omputation and randomness? What have we learned? What don’t
A Formal Theory of Creativity to Model the Creation of Art
According to the Formal Theory of Creativity (1990–2010), a creative agent—one that never stops generating non-trivial, novel, and surprising behaviours and data—must have two learning components: a
Maximizing Fun by Creating Data with Easily Reducible Subjective Complexity
  • J. Schmidhuber
  • Computer Science
    Intrinsically Motivated Learning in Natural and Artificial Systems
  • 2013
The Formal Theory of Fun and Creativity (1990–2010) describes principles of a curious and creative agent that never stops generating nontrivial & novel & surprising tasks and data.


Gödel Machines: Towards a Technical Justification of Consciousness
  • J. Schmidhuber
  • Computer Science
    Adaptive Agents and Multi-Agent Systems
  • 2005
Here it is pointed out that the recently formulated Godel machines may provide just such a technical justification for the usefulness of consciousness, and are the first mathematically rigorous, general, fully self-referential, self-improving, optimally efficient problem solvers.
Gödel Machines: Fully Self-referential Optimal Universal Self-improvers
The first class of mathematically rigorous, general, fully self-referential, self-improving, optimally efficient problem solvers is presented, which not only boasts an optimal order of complexity but can optimally reduce any slowdowns hidden by the O()-notation, provided the utility of such speed-ups is provable at all.
A Computer Scientist's View of Life, the Universe, and Everything
Basic concepts of Kolmogorov complexity theory are applied to the set of possible universes, and perceived and true randomness, life, generalization, and learning in a given universe are discussed.
Completely Self-referential Optimal Reinforcement Learners
This work presents the first class of mathematically rigorous, general, fully self-referential, self-improving, optimal reinforcement learning systems, which not only boasts an optimal order of complexity but can optimally reduce any slowdowns hidden by the O()- notation, provided the utility of such speed-ups is provable at all.
Algorithmic Theories of Everything
We make the plausible assumption that the history of our universe is formally describable, and sampled from a formally describable probability distribution on the possible universe histories. To
Computers and Thought
Computers and Thought showcases the work of the scientists who not only defined the field of Artificial Intelligence, but who are responsible for having developed it into what it is today. Originally
Hierarchies of Generalized Kolmogorov Complexities and Nonenumerable Universal Measures Computable in the Limit
  • J. Schmidhuber
  • Mathematics, Computer Science
    Int. J. Found. Comput. Sci.
  • 2002
A natural hierarchy of generalizations of algorithmic probability and Kolmogorov complexity is obtained, suggesting that the "true" information content of some bitstring x is the size of the shortest nonhalting program that converges to x and nothing but x on a Turing machine that can edit its previous outputs.
Complexity-based induction systems: Comparisons and convergence theorems
Levin has shown that if tilde{P}'_{M}(x) is an unnormalized form of this measure, and P( x) is any computable probability measure on strings, x, then \tilde{M}'_M}\geqCP (x) where C is a constant independent of x .
Towards a Universal Theory of Artificial Intelligence Based on Algorithmic Probability and Sequential Decisions
This work unifies both theories of universal induction and gives strong arguments that the resulting universal AIξ model behaves optimally in any computable environment, which is still superior to any other time t and length l bounded agent.