Algorithmic Iteration for Computational Intelligence
Machine awareness is a disputed research topic, in some circles considered a crucial step in realising Artificial General Intelligence. Understanding what that is, under which conditions such feature could arise and how it can be controlled is still a matter of speculation. A more concrete object of theoretical analysis is algorithmic iteration for computational intelligence, intended as the theoretical and practical ability of algorithms to design other algorithms for actions aimed at solving well-specified tasks. We know this ability is already shown by current AIs, and understanding its limits is an essential step in qualifying claims about machine awareness and Super-AI. We propose a formal translation of algorithmic iteration in a fragment of modal logic, formulate principles of transparency and faithfulness across human and machine intelligence, and consider the relevance to theoretical research on (Super)-AI as well as the practical import of our results.
KeywordsArtificial intelligence Introspection Machine awareness Algorithm design Algorithm execution
Thanks to Raf Pasmans, who unexpectedly prompted my first thoughts on this topic. Thanks to the participants at the ‘Debate on Moral and Philosophical Aspects of Artificial Intelligence’ at Ghent University and at the ‘Conference on Artificial Intelligence and Contemporary Society: The Role of Information’ at University of A Coruña, where early versions of this paper were presented. I am especially grateful to Wenceslao Gonzalez, Luciano Floridi, Patrick Allo and Nikos Gkorogiannis, whose observations and comments helped address some conceptual and formal issues. Two anonymous referees’ critiques have helped improving the final version of this paper.
- Adelson, B., & Soloway, E. (2007). A model of software design. International Journal of Intelligent Systems, 1(3), 195–213, 1986 (republished).Google Scholar
- Armstrong, D. M. (1981). The nature of mind and other essays. Ithaca: Cornell University Press.Google Scholar
- Barstow, D. (1984). A perspective on automatic programming. AI Magazine, 5(1), 5–27.Google Scholar
- Bolander, T. (2003). Logical theories for agent introspection. Ph.D. thesis, informatics and mathematical modelling, Technical University of Denmark.Google Scholar
- Bolander, T. (2015). Self-reference. The Stanford encyclopedia of philosophy. In E. N. Zalta (Ed.), Spring Edition. http://plato.stanford.edu/archives/spr2015/entries/self-reference/.
- Bostrom, N. (2014). Superintelligence: Paths, dangers, strategies. New York: Oxford University Press.Google Scholar
- Chalmers, D. (1996). The conscious mind: In search of a fundamental theory. Oxford: Oxford University Press.Google Scholar
- Daylight, E. G. (2015). Towards a historical notion of ‘Turing—The Father of Computer Science’. History and Philosophy of Logic, 36(3), 205–228. http://www.tandfonline.com/doi/full/10.1080/01445340.2015.1082050.
- De Bruijn, N. (1983). Automath, a language for mathematics. Department of Mathematics, Eindhoven University of Technology, TH-report 68-WSK-05, 1968. Reprinted in revised form, with two pages commentary. In Automation and Reasoning, vol. 2, Classical papers on computational logic 1967–1970, pp. 159–200. Berlin: Springer.Google Scholar
- De Grave, K. (Ed.) (2015). Formalism and intuition in software development. A conversation with Michael A. Jackson conducted by Edgar G. Daylight and Bas van Vlijmen. Conversations, Issue 5, Lonely Scholar.Google Scholar
- Evans, G. (1982). The varieties of reference. Oxford: Oxford University Press.Google Scholar
- Fallenstein, B., & Soares, N. (2015). Vingean reflection: Reliable reasoning for self-improving agents. Technical Report 2015-2, Machine Intelligence Research Institute.Google Scholar
- Fischer, M. J., & Ladner, R. E. (1977). Propositional modal logic of programs. In STOC ’77 Proceedings of the ninth annual ACM symposium on theory of computing, pp. 286–294 .Google Scholar
- Floridi, L. (2015). Singularitarians, aitheists, and why the problem with artificial intelligence is H.A.L. (humanity at large), not HAL. APA Newsletter, 14(2), 7–11.Google Scholar
- Gertler, B. Self-knowledge. The Stanford Encyclopedia of Philosophy (Summer 2015 Edition). In E. N. Zalta (Ed.) http://plato.stanford.edu/archives/sum2015/entries/self-knowledge/.
- Goel, A. K., Morse, E. L., Raja, A., Scholtz, J., & Stasko, J. T. (2009). Computational explanations for report generation in intelligence analysis. ExaCt, 37–47, 2009.Google Scholar
- Good, I. J. (1965) Speculations concerning the First ultraintelligent machine. In F. L. Alt and M. Rubinoff (eds.), Advances in computers (Vol. 6, pp. 31–88).Google Scholar
- Gurevich, Y. (2012). What is an algorithm? SOFSEM 2012: Theory and practice of computer science. Lecture Notes in Computer Science, (Vol. 7147, pp. 31–42).Google Scholar
- Halpern, J. Y., & Vardi, M. (1994). Algorithmic knowledge. In R. Fagin (Ed.), Proceedings of the 5th conference on theoretical aspects of reasoning about knowledge (pp. 255–266). Morgan Kaufmann.Google Scholar
- Hibbard, B. (2012). Decision support for safe AI design. In J. Bach, B. Goertzel and M. Iklé (Eds.), Artificial general intelligence. Lecture notes in artificial intelligence (Vol. 7716, pp. 117–25). New York: Springer.Google Scholar
- Howard, W. A. (1980). The formulae-as-types notion of construction. In Seldin, Jonathan P., Hindley, J. Roger, To H.B. Curry: Essays on combinatory logic, lambda calculus and formalism (pp. 479–490). Boston, MA: Academic Press (original paper manuscript from 1969).Google Scholar
- Jeffries, R., Turner, A. A., & Polson, P. G. (1981). The processes involved in designing software. In J. R. Anderson (Ed.), Cognitive skills and their acquisition, ch. 8. Hillsdale, NJ: Lawrence Erlbaum.Google Scholar
- Kant, E. (1985). Understanding and automating algorithm design. IEEE Transactions on Software Engineering, SE-11, 1243–1253.Google Scholar
- Kleinberg, J., & Tardos, E. (2005). Algorithm design. Reading, MA: Pearson Addison-Wesley.Google Scholar
- Konolige, K. (1985). A computational theory of belief introspection. IJCAI, 85, 503–508.Google Scholar
- van Leeuwen, J. J., & Wiedermann, J. (2001). Beyond the Turing limit: Evolving interactive systems. In L. Pacholski, P. Ruzicka (Eds.), SOFSEM, in Lecture Notes in Computer Science (vol. 2234, pp. 90–109).Google Scholar
- McKay, T., Nelson, M. (2014). Propositional attitude reports. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Spring Edition). http://plato.stanford.edu/archives/spr2014/entries/prop-attitude-reports/.
- Menzel, C. A. (2016). The Stanford encyclopedia of philosophy (Summer Edition), E. N. Zalta (ed.). http://plato.stanford.edu/archives/sum2016/entries/actualism/.
- Moschovakis, Y. N. (1994). Sense and denotation as algorithm and value. In J. Oikkonen and J. Vaananen (Eds.), Lecture notes in logic (Vol. 2, pp. 210–249). Berlin: SpringerGoogle Scholar
- Moschovakis, Y. N. (2001). What is an algorithm? In B. Engquist and W. Schmid (Eds.), Mathematics unlimited—2001 and beyond (pp. 919–936). Berlin: Springer.Google Scholar
- Pratt, V. R. (1976) Semantical considerations on Floyd-Hoare Logic. In SFCS ’76 proceedings of the 17th annual symposium on foundations of computer science, pp. 109–121Google Scholar
- Primiero, G. (2015). Realist consequence, epistemic inference, computational correctness. In A. Koslow & A. Buchsbaum (Eds.), The road to universal logic, Part of the series Studies in Universal Logic (Vol. 2, pp. 573–588). Springer, Birkhauser.Google Scholar
- Primiero, G. (2016). Information in the philosophy of computer science. In L. Floridi (Ed.), The Routledge handbook of philosophy of information, ch. 10 (pp. 90–106).Google Scholar
- Schwitzgebel, E. (2014). Introspection. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Summer Edition). http://plato.stanford.edu/archives/sum2014/entries/introspection/.
- Shoemaker, S. (1994). Self-knowledge and ‘Inner-Sense’. Philosophy and Phenomenological Research, 54, 249–314. Reprinted in The First Person Perspective and other Essays, OUP, 1996.Google Scholar
- Sonntag, D. (2008). On introspection, metacognitive control and augmented data mining live cycles. 0807.4417[CoRRabs].Google Scholar
- Sørensen, M. H., & Urzyczyn, P. (2006). Lectures on the Curry-Howard isomorphism. Studies in Logic and the Foundations of Mathematics (Vol. 149). London: Elsevier.Google Scholar
- Sotala, K., Yampolskiy, R. (2015). Responses to catastrophic AGI risk: A survey. Physica Scripta, 90. Royal Swedish Academy of Sciences, IOP.Google Scholar
- van Ditmarsch, H., & French, T. (2011). On the interactions of awareness and certainty. Australasian Conference on Artificial Intelligence, pp. 727–738.Google Scholar
- Voss, P. (2007). Essentials of general intelligence: The direct path to artificial general intelligence In B. Goertzel & C. Pennachin (Eds.), Artificial General Intelligence (pp. 131–157). Heidelberg: Springer.Google Scholar