Rationality and Intelligence: A Brief Update

  • Stuart RussellEmail author
Part of the Synthese Library book series (SYLI, volume 376)


The long-term goal of AI is the creation and understanding of intelligence. This requires a notion of intelligence that is precise enough to allow the cumulative development of robust systems and general results. The concept of rational agency has long been considered a leading candidate to fulfill this role. This paper, which updates a much earlier version (Russell, Artif Intell 94:57–77, 1997), reviews the sequence of conceptual shifts leading to a different candidate, bounded optimality, that is closer to our informal conception of intelligence and reduces the gap between theory and practice. Some promising recent developments are also described.


Rationality Intelligence Bounded rationality Metareasoning 



An earlier version of this paper appeared in the journal Artificial Intelligence, published by Elsevier. That paper drew on previous work with Eric Wefald and Devika Subramanian. More recent results were obtained with Nick Hay. Thanks also to Michael Wellman, Michael Fehling, Michael Genesereth, Russ Greiner, Eric Horvitz, Henry Kautz, Daphne Koller, Bart Selman, and Daishi Harada for many stimulating discussions topic of bounded rationality. The research was supported by NSF grants IRI-8903146, IRI-9211512 and IRI-9058427, and by a UK SERC Visiting Fellowship. The author is supported by the Chaire Blaise Pascal, funded by the l’État et la Région Île de France and administered by the Fondation de l’École Normale Supérieure.


  1. Agre, P. E., & Chapman, D. (1987). Pengi: An implementation of a theory of activity. In Proceedings of the Tenth International Joint Conference on Artificial Intelligence (IJCAI-87), Milan (pp. 268–272). Morgan Kaufmann.Google Scholar
  2. Andre, D., & Russell, S. J. (2002) State abstraction for programmable reinforcement learning agents. In Proceedings of the Eighteenth National Conference on Artificial Intelligence (AAAI-02), Edmonton (pp. 119–125). AAAI Press.Google Scholar
  3. Bellman, R. E. (1957). Dynamic programming. Princeton: Princeton University Press.Google Scholar
  4. Berry, D. A., & Fristedt, B. (1985). Bandit problems: Sequential allocation of experiments. London: Chapman and Hall.CrossRefGoogle Scholar
  5. Breese, J. S., & Fehling, M. R. (1990). Control of problem-solving: Principles and architecture. In R. D. Shachter, T. Levitt, L. Kanal, & J. Lemmer (Eds.), Uncertainty in artificial intelligence 4. Amsterdam/London/New York: Elsevier/North-Holland.Google Scholar
  6. Brooks, R. A. (1989). Engineering approach to building complete, intelligent beings. Proceedings of the SPIE—The International Society for Optical Engineering, 1002, 618–625.Google Scholar
  7. Carnap, R. (1950). Logical foundations of probability. Chicago: University of Chicago Press.Google Scholar
  8. Cherniak, C. (1986). Minimal rationality. Cambridge: MIT.Google Scholar
  9. Dean, T., & Boddy, M. (1988) An analysis of time-dependent planning. In Proceedings of the Seventh National Conference on Artificial Intelligence (AAAI-88), St. Paul (pp. 49–54). Morgan Kaufmann.Google Scholar
  10. Dean, T., Aloimonos, J., & Allen, J. F. (1995). Artificial intelligence: Theory and practice. Redwood City: Benjamin/Cummings.Google Scholar
  11. Dennett, D. C. (1988). The moral first aid manual. In S. McMurrin (Ed.), Tanner lectures on human values (Vol. 7, pp. 121–147). University of Utah Press and Cambridge University Press.Google Scholar
  12. Doyle, J., & Patil, R. (1991). Two theses of knowledge representation: Language restrictions, taxonomic classification, and the utility of representation services. Artificial Intelligence, 48(3), 261–297CrossRefGoogle Scholar
  13. Good, I. J. (1971) Twenty-seven principles of rationality. In V. P. Godambe & D. A. Sprott (Eds.), Foundations of statistical inference (pp. 108–141). Toronto: Holt, Rinehart, Winston.Google Scholar
  14. Goodman, N. D., Mansinghka, V. K., Roy, D. M., Bonawitz, K., & Tenenbaum, J. B. (2008). Church: A language for generative models. In Proceedings of UAI-08, Helsinki (pp. 220–229).Google Scholar
  15. Harman, G. H. (1983). Change in view: Principles of reasoning. Cambridge: MIT.Google Scholar
  16. Hay, N., Russell, S., Shimony, S. E., & Tolpin, D. (2012). Selecting computations: Theory and applications. In Proceedings of UAI-12, Catalina Island.Google Scholar
  17. Horvitz, E. J. (1987). Problem-solving design: Reasoning about computational value, trade-offs, and resources. In Proceedings of the Second Annual NASA Research Forum, NASA Ames Research Center, Moffett Field, CA (pp. 26–43).Google Scholar
  18. Horvitz, E. J. (1989). Reasoning about beliefs and actions under computational resource constraints. In L. N. Kanal, T. S. Levitt, & J. F. Lemmer (Eds.), Uncertainty in artificial intelligence 3 (pp. 301–324). Amsterdam/London/New York: Elsevier/North-Holland.Google Scholar
  19. Horvitz, E. J., & Breese, J. S. (1990). Ideal partition of resources for metareasoning (Technical report KSL-90-26), Knowledge Systems Laboratory, Stanford University, Stanford.Google Scholar
  20. Howard, R. A. (1966). Information value theory. IEEE Transactions on Systems Science and Cybernetics, SSC-2, 22–26.CrossRefGoogle Scholar
  21. Hutter, M. (2005). Universal artificial intelligence: Sequential decisions based on algorithmic probability. Berlin/New York: Springer.Google Scholar
  22. Kearns, M., Schapire, R. E., & Sellie, L. (1992). Toward efficient agnostic learning. In Proceedings of the Fifth Annual ACM Workshop on Computational Learning Theory (COLT-92), Pittsburgh. ACM.Google Scholar
  23. Keeney, R. L., & Raiffa, H. (1976). Decisions with multiple objectives: Preferences and value tradeoffs. New York: Wiley.Google Scholar
  24. Kocsis, L., & Szepesvari, C. (2006). Bandit-based Monte-Carlo planning. In Proceedings of ECML-06, Berlin.Google Scholar
  25. Kolmogorov, A. N. (1965). Three approaches to the quantitative definition of information. Problems in Information Transmission, 1(1), 1–7.Google Scholar
  26. Koopmans, T. C. (1972). Representation of preference orderings over time. In C.B. McGuire & R. Radner (Eds.), Decision and organization. Amsterdam/London/New York: Elsevier/North-Holland.Google Scholar
  27. Kumar, P. R., & Varaiya, P. (1986). Stochastic systems: Estimation, identification, and adaptive control. Upper Saddle River: Prentice-Hall.Google Scholar
  28. Laird, J. E., Rosenbloom, P. S., & Newell, A. (1986). Chunking in Soar: The anatomy of a general learning mechanism. Machine Learning, 1, 11–46.Google Scholar
  29. Levesque, H. J. (1986). Making believers out of computers. Artificial Intelligence, 30(1), 81–108.CrossRefGoogle Scholar
  30. Livnat, A., & Pippenger, N. (2006). An optimal brain can be composed of conflicting agents. Proceedings of the National Academy of Sciences of the United States of America 103(9), 3198–3202.CrossRefGoogle Scholar
  31. Marthi, B., Russell, S., Latham, D., & Guestrin, C. (2005). Concurrent hierarchical reinforcement learning. In Proceedings of IJCAI-05, Edinburgh.Google Scholar
  32. Marthi, B., Russell, S. J., & Wolfe, J. (2008). Angelic hierarchical planning: Optimal and online algorithms. In Proceedings of ICAPS-08, Sydney.Google Scholar
  33. Matheson, J. E. (1968). The economic value of analysis and computation. IEEE Transactions on Systems Science and Cybernetics, SSC-4(3), 325–332.CrossRefGoogle Scholar
  34. Megiddo, N., & Wigderson, A. (1986). On play by means of computing machines. In J. Y. Halpern (Ed.), Theoretical Aspects of Reasoning About Knowledge: Proceedings of the 1986 Conference (TARK-86), IBM and AAAI, Monterey (pp. 259–274). Morgan Kaufmann.Google Scholar
  35. Milch, B., Marthi, B., Sontag, D., Russell, S. J., Ong, D., & Kolobov, A. (2005). BLOG: Probabilistic models with unknown objects. In Proceedings of IJCAI-05, Edinburgh.Google Scholar
  36. Newell, A. (1982). The knowledge level. Artificial Intelligence, 18(1), 82–127.CrossRefGoogle Scholar
  37. Nilsson, N. J. (1991). Logic and artificial intelligence. Artificial Intelligence, 47(1–3), 31–56CrossRefGoogle Scholar
  38. Papadimitriou, C. H., & Yannakakis, M. (1994). On complexity as bounded rationality. In Symposium on Theory of Computation (STOC-94), Montreal.Google Scholar
  39. Parr, R., & Russell, S. J. (1998). Reinforcement learning with hierarchies of machines. In M. I. Jordan, M. Kearns, & S. A. Solla (Eds.), Advances in neural information processing systems 10. Cambridge: MIT.Google Scholar
  40. Pfeffer, A. (2001). IBAL: A probabilistic rational programming language. In Proceedings of IJCAI-01, Seattle (pp. 733–740).Google Scholar
  41. Russell, S. J. (1997). Rationality and intelligence. Artificial Intelligence, 94, 57–77.CrossRefGoogle Scholar
  42. Russell, S. J. (1998). Learning agents for uncertain environments (extended abstract). In Proceedings of the Eleventh Annual ACM Workshop on Computational Learning Theory (COLT-98), Madison (pp. 101–103). ACM.Google Scholar
  43. Russell, S. J., & Norvig, P. (1995). Artificial intelligence: A modern approach. Upper Saddle River: Prentice-Hall.Google Scholar
  44. Russell, S. J., & Subramanian, D. (1995). Provably bounded-optimal agents. Journal of Artificial Intelligence Research, 3, 575–609.Google Scholar
  45. Russell, S. J., & Wefald, E. H. (1989). On optimal game-tree search using rational meta-reasoning. In Proceedings of the Eleventh International Joint Conference on Artificial Intelligence (IJCAI-89), Detroit (pp. 334–340). Morgan Kaufmann.Google Scholar
  46. Russell, S. J., & Wefald, E. H. (1991a). Do the right thing: Studies in limited rationality. Cambridge: MIT.Google Scholar
  47. Russell, S. J., & Wefald, E. H. (1991b). Principles of metareasoning. Artificial Intelligence 49(1–3), 361–395.Google Scholar
  48. Russell, S. J., & Zilberstein, S. (1991). Composing real-time systems. In Proceedings of the Twelfth International Joint Conference on Artificial Intelligence (IJCAI-91), Sydney. Morgan Kaufmann.Google Scholar
  49. Shoham, Y., & Leyton-Brown, K. (2009). Multiagent systems: Algorithmic, game-theoretic, and logical foundations. Cambridge/New York: Cambridge University Press.Google Scholar
  50. Simon, H. A. (1955). A behavioral model of rational choice. Quarterly Journal of Economics, 69, 99–118.CrossRefGoogle Scholar
  51. Simon, H. A. (1958). Rational choice and the structure of the environment. In Models of bounded rationality (Vol. 2). Cambridge: MIT.Google Scholar
  52. Solomonoff, R. J. (1964). A formal theory of inductive inference. Information and Control, 7, 1–22, 224–254.Google Scholar
  53. Srivastava, S., Russell, S., Ruan, P., & Cheng, X. (2014). First-order open-universe POMDPs. In Proceedings of UAI-14, Quebec City.Google Scholar
  54. Sutton, R., Precup, D., & Singh, S. P. (1999). Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning. Artificial Intelligence, 112, 181–211.CrossRefGoogle Scholar
  55. Tadepalli, P. (1991). A formalization of explanation-based macro-operator learning. In Proceedings of the Twelfth International Joint Conference on Artificial Intelligence (IJCAI-91), Sydney (pp. 616–622). Morgan Kaufmann.Google Scholar
  56. Tennenholtz, M. (2004). Program equilibrium. Games and Economic Behavior, 49(2), 363–373.CrossRefGoogle Scholar
  57. Vapnik, V. (2000). The nature of statistical learning theory. Berlin/New York: Springer.CrossRefGoogle Scholar
  58. von Neumann, J., & Morgenstern, O. (1944). Theory of games and economic behavior (1st ed.). Princeton: Princeton University Press.Google Scholar
  59. Wellman, M. P. (1994). A market-oriented programming environment and its application to distributed multicommodity flow problems. Journal of Artificial Intelligence Research, 1(1), 1–23.Google Scholar
  60. Wellman, M. P., & Doyle, J. (1991). Preferential semantics for goals. In Proceedings of the Ninth National Conference on Artificial Intelligence (AAAI-91), Anaheim (Vol. 2, pp. 698–703). AAAI Press.Google Scholar
  61. Zilberstein, S., & Russell, S. J. (1996). Optimal composition of real-time systems. Artificial Intelligence 83, 181–213.CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.Computer Science DivisionUniversity of CaliforniaBerkeleyUSA

Personalised recommendations