Beyond the Turing Test

@article{HernndezOrallo2000BeyondTT,
  title={Beyond the Turing Test},
  author={Jos{\'e} Hern{\'a}ndez-Orallo},
  journal={Journal of Logic, Language and Information},
  year={2000},
  volume={9},
  pages={447-466}
}
  • J. Hernández-Orallo
  • Published 1 October 2000
  • Computer Science
  • Journal of Logic, Language and Information
The main factor of intelligence is defined as the ability tocomprehend, formalising this ability with the help of new constructsbased on descriptional complexity. The result is a comprehension test,or C-test, which is exclusively defined in computational terms. Due toits absolute and non-anthropomorphic character, it is equally applicableto both humans and non-humans. Moreover, it correlates with classicalpsychometric tests, thus establishing the first firm connection betweeninformation… 
Turing Tests with Turing Machines
TLDR
This paper brings the Turing Test to the realm of Turing machines and discusses what it would be like to have a Turing Test where the reference and the interrogator subjects are replaced by Turing Machines.
How universal can an intelligence test be?
TLDR
It is argued that such tests must be highly adaptive, i.e. that tasks, resolution, rewards and communication have to be adapted according to how the evaluated agent is reacting and performing, and set the quest for universal tests as a progressive rather than absolute goal.
Turing Test Revisited: A Framework for an Alternative
TLDR
The paper presents a plausible generic framework based on categories of factors implied by subjective perception of intelligence, which is followed in analysing what is needed to devise a test or tests for intelligent machines.
The Measure of All Minds: Evaluating Natural and Artificial Intelligence
TLDR
Using algorithmic information theory as a foundation, the book elaborates on the evaluation of perceptual, developmental, social, verbal and collective features and critically analyzes what the future of intelligence might look like.
Building Thinking Machines by Solving Animal Cognition Tasks
TLDR
It is argued that to fix shortcomings with modern AI systems a nonverbal operationalisation is required and this is provided by the recent Animal-AI Testbed, which translates animal cognition tests for AI and provides a bottom-up research pathway for building thinking machines that create predictive models of their environment from sensory input.
Universal psychometrics: Measuring cognitive abilities in the machine kingdom
Evaluation in artificial intelligence: from task-oriented to ability-oriented measurement
TLDR
This paper critically assess the different ways AI systems are evaluated, and the role of components and techniques in these systems, and identifies three kinds of evaluation: human discrimination, problem benchmarks and peer confrontation.
An Overview of Approaches Evaluating Intelligence of Artificial Systems
TLDR
The Algorithmic Intelligence Quotient test, derived from Universal Intelligence definition, is currently the most suitable candidate for a practical intelligence evaluation method of artificial systems, although the test has several known limitations.
...
...

References

SHOWING 1-10 OF 90 REFERENCES
A formalization of the Turing test
TLDR
If human intelligence subsumes machine intelligence, and human intelligence is not simulatable by any bounded machine, then the Turing Test can distinguish humans and machines to within arbitrarily high probability, according to the notion of completeness for standard complexity classes.
Machines and Thought
TLDR
This book discusses subcognition and the Limits of the Turing Test, Turing's Test and Conscious Thought, and a Grammar-Based Approach to Common-Sense Reasoning.
Lessons from a restricted Turing test
TLDR
It is argued that the Loebner prize competition has no clear purpose, that its design prevents any useful outcome, and that such a competition is inappropriate given the current level of technology.
The Turing Test is not a trick: Turing indistinguishability is a scientific criterion
It is important to understand that the Turing Test (TT) is not, nor was it intended to be, a trick; how well one can fool someone is not a measure of scientific progress. The TT is an empirical
THE TURING TEST MISUNDERSTOOD
The Turing test was conceived by Alan Turing [1] as a device to get around the difficult question of whether computers can think, and replace it with one which would be easier to discuss and answer.
Explanatory and Creative Alternatives to the MDL priciple
TLDR
The Minimum Description Length principle has been extensively and successfully used in machine learning (ML), especially for noisy and long sources of data, but it presents some paradoxes andiences.
The universal turing machine: a half-century survey
  • P. Kidwell
  • History
    IEEE Annals of the History of Computing
  • 1996
TLDR
A large number of repeated myths and oversimplifications that intersperse the text should have been able to spot the former easily, and a scholar of computing would eliminate the rest.
Thinking Machines: Can There Be? Are We?
TLDR
The guiding principles of artificial intelligence are reviewed and it is argued that as now conceived it is limited to a very particular kind of intelligence: one that can usefully be likened to bureaucracy.
Handbook of Intelligence: Theories, Measurements, and Applications
THEORIES: Genetic Determinants of Intelligence Cognitive Approaches to Intelligence Neurological Foundations of Intelligence Genetic Epistemology and the Development of Intelligence General
...
...