The ANYNT project intelligence test Λone

Javier Insa-Cabrera, Jose Hernandez-Orallo, David L. Dowe, Sergio Espana, M. Victoria Hernandez-Lloreda

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    5 Citations (Scopus)

    Abstract

    All tests in psychometrics, comparative psychology and cognition which have been put into practice lack a mathematical (computational) foundation or lack the capability to be applied to any kind of system (humans, non-human animals, machines, hybrids, collectives, etc.). In fact, most of them lack both things. In the past fifteen years, some efforts have been done to derive intelligence tests from formal intelligence definitions or vice versa, grounded on computational concepts. However, some of these approaches have not been able to create universal tests (i.e., tests which can evaluate any kind of subjects) and others have even failed to make a feasible test. The ANYNT project was conceived to explore the possibility of defining formal, universal and anytime intelligence tests, having a feasible implementation in mind. This paper presents the basics of the theory behind the ANYNT project and describes one of the test propotypes that were developed in the project: test Λone.
    Original languageEnglish
    Title of host publicationAISB/IACAP World Congress 2012: Revisiting Turing and his Test: Comprehensiveness, Qualia, and the Real World, Part of Alan Turing Year 2012
    Subtitle of host publication2 - 6 July 2012, Birmingham, United Kingdom
    EditorsVincent C Muller, Aladdin Ayesh
    Place of PublicationHove, UK
    PublisherSociety for the Study of Artificial Intelligence and Simulation of Behaviour
    Pages20 - 27
    Number of pages8
    ISBN (Print)9781908187154
    Publication statusPublished - 2012
    EventAISB Convention (Society for the Study of Artificial Intelligence and Simulation of Behaviour) 2012 - University of Birmingham, Birmingham, United Kingdom
    Duration: 2 Jul 20126 Jul 2012
    http://events.cs.bham.ac.uk/turing12/

    Conference

    ConferenceAISB Convention (Society for the Study of Artificial Intelligence and Simulation of Behaviour) 2012
    Abbreviated titleAISB 2012
    CountryUnited Kingdom
    CityBirmingham
    Period2/07/126/07/12
    OtherFor the Turing year 2012, the AISB (The Society for the Study of Artificial Intelligence and Simulation of Behaviour) and the IACAP (The International Association for Computing and Philosophy) have merged their annual symposia/conferences to form the AISB/IACAP World Congress. The congress will take place 2-6 July 2012 in Birmingham, UK.
    Internet address

    Keywords

    • (Machine) intelligence evaluation
    • Artificial intelligence
    • Solomonoff-Kolmogorov complexity
    • Universal tests

    Cite this

    Insa-Cabrera, J., Hernandez-Orallo, J., Dowe, D. L., Espana, S., & Hernandez-Lloreda, M. V. (2012). The ANYNT project intelligence test Λone. In V. C. Muller, & A. Ayesh (Eds.), AISB/IACAP World Congress 2012: Revisiting Turing and his Test: Comprehensiveness, Qualia, and the Real World, Part of Alan Turing Year 2012: 2 - 6 July 2012, Birmingham, United Kingdom (pp. 20 - 27). Hove, UK: Society for the Study of Artificial Intelligence and Simulation of Behaviour.
    Insa-Cabrera, Javier ; Hernandez-Orallo, Jose ; Dowe, David L. ; Espana, Sergio ; Hernandez-Lloreda, M. Victoria. / The ANYNT project intelligence test Λone. AISB/IACAP World Congress 2012: Revisiting Turing and his Test: Comprehensiveness, Qualia, and the Real World, Part of Alan Turing Year 2012: 2 - 6 July 2012, Birmingham, United Kingdom. editor / Vincent C Muller ; Aladdin Ayesh. Hove, UK : Society for the Study of Artificial Intelligence and Simulation of Behaviour, 2012. pp. 20 - 27
    @inproceedings{87c473859898475e9ce06e2c3a7b0c65,
    title = "The ANYNT project intelligence test Λone",
    abstract = "All tests in psychometrics, comparative psychology and cognition which have been put into practice lack a mathematical (computational) foundation or lack the capability to be applied to any kind of system (humans, non-human animals, machines, hybrids, collectives, etc.). In fact, most of them lack both things. In the past fifteen years, some efforts have been done to derive intelligence tests from formal intelligence definitions or vice versa, grounded on computational concepts. However, some of these approaches have not been able to create universal tests (i.e., tests which can evaluate any kind of subjects) and others have even failed to make a feasible test. The ANYNT project was conceived to explore the possibility of defining formal, universal and anytime intelligence tests, having a feasible implementation in mind. This paper presents the basics of the theory behind the ANYNT project and describes one of the test propotypes that were developed in the project: test Λone.",
    keywords = "(Machine) intelligence evaluation, Artificial intelligence, Solomonoff-Kolmogorov complexity, Universal tests",
    author = "Javier Insa-Cabrera and Jose Hernandez-Orallo and Dowe, {David L.} and Sergio Espana and Hernandez-Lloreda, {M. Victoria}",
    year = "2012",
    language = "English",
    isbn = "9781908187154",
    pages = "20 -- 27",
    editor = "Muller, {Vincent C} and Aladdin Ayesh",
    booktitle = "AISB/IACAP World Congress 2012: Revisiting Turing and his Test: Comprehensiveness, Qualia, and the Real World, Part of Alan Turing Year 2012",
    publisher = "Society for the Study of Artificial Intelligence and Simulation of Behaviour",

    }

    Insa-Cabrera, J, Hernandez-Orallo, J, Dowe, DL, Espana, S & Hernandez-Lloreda, MV 2012, The ANYNT project intelligence test Λone. in VC Muller & A Ayesh (eds), AISB/IACAP World Congress 2012: Revisiting Turing and his Test: Comprehensiveness, Qualia, and the Real World, Part of Alan Turing Year 2012: 2 - 6 July 2012, Birmingham, United Kingdom. Society for the Study of Artificial Intelligence and Simulation of Behaviour, Hove, UK, pp. 20 - 27, AISB Convention (Society for the Study of Artificial Intelligence and Simulation of Behaviour) 2012, Birmingham, United Kingdom, 2/07/12.

    The ANYNT project intelligence test Λone. / Insa-Cabrera, Javier; Hernandez-Orallo, Jose; Dowe, David L.; Espana, Sergio; Hernandez-Lloreda, M. Victoria.

    AISB/IACAP World Congress 2012: Revisiting Turing and his Test: Comprehensiveness, Qualia, and the Real World, Part of Alan Turing Year 2012: 2 - 6 July 2012, Birmingham, United Kingdom. ed. / Vincent C Muller; Aladdin Ayesh. Hove, UK : Society for the Study of Artificial Intelligence and Simulation of Behaviour, 2012. p. 20 - 27.

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    TY - GEN

    T1 - The ANYNT project intelligence test Λone

    AU - Insa-Cabrera, Javier

    AU - Hernandez-Orallo, Jose

    AU - Dowe, David L.

    AU - Espana, Sergio

    AU - Hernandez-Lloreda, M. Victoria

    PY - 2012

    Y1 - 2012

    N2 - All tests in psychometrics, comparative psychology and cognition which have been put into practice lack a mathematical (computational) foundation or lack the capability to be applied to any kind of system (humans, non-human animals, machines, hybrids, collectives, etc.). In fact, most of them lack both things. In the past fifteen years, some efforts have been done to derive intelligence tests from formal intelligence definitions or vice versa, grounded on computational concepts. However, some of these approaches have not been able to create universal tests (i.e., tests which can evaluate any kind of subjects) and others have even failed to make a feasible test. The ANYNT project was conceived to explore the possibility of defining formal, universal and anytime intelligence tests, having a feasible implementation in mind. This paper presents the basics of the theory behind the ANYNT project and describes one of the test propotypes that were developed in the project: test Λone.

    AB - All tests in psychometrics, comparative psychology and cognition which have been put into practice lack a mathematical (computational) foundation or lack the capability to be applied to any kind of system (humans, non-human animals, machines, hybrids, collectives, etc.). In fact, most of them lack both things. In the past fifteen years, some efforts have been done to derive intelligence tests from formal intelligence definitions or vice versa, grounded on computational concepts. However, some of these approaches have not been able to create universal tests (i.e., tests which can evaluate any kind of subjects) and others have even failed to make a feasible test. The ANYNT project was conceived to explore the possibility of defining formal, universal and anytime intelligence tests, having a feasible implementation in mind. This paper presents the basics of the theory behind the ANYNT project and describes one of the test propotypes that were developed in the project: test Λone.

    KW - (Machine) intelligence evaluation

    KW - Artificial intelligence

    KW - Solomonoff-Kolmogorov complexity

    KW - Universal tests

    UR - http://www.scopus.com/inward/record.url?scp=84893100979&partnerID=8YFLogxK

    M3 - Conference Paper

    SN - 9781908187154

    SP - 20

    EP - 27

    BT - AISB/IACAP World Congress 2012: Revisiting Turing and his Test: Comprehensiveness, Qualia, and the Real World, Part of Alan Turing Year 2012

    A2 - Muller, Vincent C

    A2 - Ayesh, Aladdin

    PB - Society for the Study of Artificial Intelligence and Simulation of Behaviour

    CY - Hove, UK

    ER -

    Insa-Cabrera J, Hernandez-Orallo J, Dowe DL, Espana S, Hernandez-Lloreda MV. The ANYNT project intelligence test Λone. In Muller VC, Ayesh A, editors, AISB/IACAP World Congress 2012: Revisiting Turing and his Test: Comprehensiveness, Qualia, and the Real World, Part of Alan Turing Year 2012: 2 - 6 July 2012, Birmingham, United Kingdom. Hove, UK: Society for the Study of Artificial Intelligence and Simulation of Behaviour. 2012. p. 20 - 27