Classic AI Papers

AI Articles that have Stood the Test of Time

AITopics > Overview / History > Classic AI Papers

AITopics has collected digital versions of many classic AI papers that are difficult to find in print and lists others that have influenced the field. Some of the papers that every student of AI should be familiar with. Please contact us with suggestions of others that should be included here.

An extensive list of about 2000 publications relevant to the early history of AI was collected by Marvin Minsky and reprinted as A SELECTED DESCRIPTOR-INDEXED BIBLIOGRAPHY TO THE LITERATURE ON ARTIFICIAL INTELLIGENCE in Computers and Thought (1963).

Many seminal articles also appear in Semantic Information Processing, Edited by Marvin L. Minsky (1969). MIT Press. Although the book is out of print, permission to copy articles has not been granted.

Classic Papers in AI Prior to 1955

Ashby, W.Ross (1952). "Can a Mechanical Chess-player Outplay its Designer?,". British Journal for the Philosophy of Science, Vol. 3(9) pp. 44-57. Also reprinted in Mechanisms of Intelligence: Ashby's Writings on Cybernetics (1981) by Roger Conant. Also see the online Bibliography of W. Ross Ashby's papers.

Bush, Vannevar (1945). As We May Think (Atlantic Monthly, July 1945) -- A vision of the future in which computers assist humans in many activities.

Lovelace, A. Ada. (1843). Notes by the translator (to L.F. Menabrea's "Sketch of the analytical engine invented by Charles Babbage, Esq."). Scientific Memoirs, 3, 666-731. [Lady Lovelace's extensive notes to the major account of Babbage's mechanical computer.]

McCullough, Warren and Walter Pitts (1943). A logical calculus of the ideas immanent in nervous activity -- free abstract available. (Bulletin of Mathematical Biophysics, 7:115 - 133). McCullogh and Pitts, published the first paper describing what we would call a neural network. While their work did not demonstrate a system capable of learning, it did show the ability of simple neuron configurations to perform computations.

Shannon, Claude (1948). A Mathematical Theory of Communication. Bell System Technical Journal 27: 379-423 and 623-656.

Shannon, Claude (1950). "Programming a computer to play chess" .

Simon, H.A. (1940s-60s). Nobel Prize Acceptance Lecture. A readable summary of Simon's work on organizational theory in the 1940s, '50s and '60s for which he was awarded the Nobel Prize in 1978. "[the concept of] “bounded rationality”, [refers to the] machinery for coping with the limits of man’s abilities to comprehend and compute in the face of complexity and uncertainty. …Two concepts are central to the characterization: search and satisficing. If the alternatives for choice are not given initially to the decision maker, then he must search for them. … one could postulate that the decision maker had formed some aspiration as to how good an alternative he should find. As soon as he discovered an alternative for choice meeting his level of aspiration, he would terminate the search and choose that alternative. I called this mode of selection satisficing."

Turing, A.M. (1950). Turing: Computing Machinery and Intelligence. By Alan M. Turing (1950). Mind 59 (Oct 1950): 433-60. ["Originally published by Oxford University Press on behalf of MIND (the Journal of the Mind Association), vol. LIX, no. 236, pp. 433-60, 1950. Published on the abelard site by permission of Oxford University Press."] An all-time classic paper that discusses the prospects of AI and dismisses some still-current arguments against AI. Introduction of the Turing Test as a way of operationalizing a test of intelligent behavior. (PDF file of the orignal journal article downloadable from Oxford University Press).

Wells , H.G. (1937). "World Brain: The Idea of a Permanent World Encyclopedia". An early vision of a "world synthesis of bibliography and documentation with the indexed archives of the world" -- based on microfilm, but nevertheless anticipating the distribution of knowledge on the web.

Classic Papers in AI 1955-1974

Armer , Paul (1960). Attitudes toward intelligent machines. Symposium on Bionics, 1960, Rand Technical Report 60 600, pp. 13--19. A very readable summary of objections to the concept of machine intelligence. (Reprinted in Computers and Thought.)

Clarkson , Geoffrey P. E. (1961). A MODEL OF THE TRUST INVESTMENT PROCESS . From A Simulation of Trust Investment, Englewood Cliffs, N.J.: Prentice-Hall, 1961. (Reprinted in Computers and Thought.)

Dineen, G.P. (1955). Programming Pattern Recognition. 1955 WESTERN JOINT COMPUTER CONFERENCE. LOS ANGELES March 01-March 03, 1955. From IEEE Digitial Library.

Dreyfus, Hubert (1965). Alchemy and Artificial Intelligence . This is the first of many highly critical articles by Hubert Dreyfus, a philosopher who believed that AI could never succeed in producing programs capable of exhibiting human-level intelligence.

Feigenbaum, E.A. (1961). The simulation of verbal learning behavior. Proceedings of the Western Joint Computer Conference, 1961, 19:121-132. An early demonstration of the power of computer simulations for modeling the cognitive behavior of humans. Among other things, EPAM contains a model of forgetting. (Reprinted in Computers and Thought.)

Feigenbaum, E.A., Bruce G. Buchanan, and Joshua Lederberg (1970). Generality and Problem Solving: A Case Study Using the Dendral Program. Downloadable PDF. Machine Intelligence 6 (1970) Volume: 6, Publisher: Edinburg University Press, Pages: 165-190. A seminal paper that defined a new direction for AI in building knowledge-based expert systems. "In discussing the capability of a problem solving system, one should distinguish between generality anal expertness. Generality is being questioned when-we ask: how broad a universe of problems is the problem solver prepared to work on? Expertnness is being questioned when we ask: how good are the answers and were they arrived at with reasonable cost? Generality has great utility in some ways, but is not often associated with superior performance. The experts usually are specialists."

Fikes , Richard and Nils Nilsson (1971). STRIPS: A New Approach to the Application of Theorem Proving to Problem Solving. (Artificial Intelligence, Vol. 2, No. 3-4, pp. 189-208). The STRIPS system uses a resolution theorem prover to answer questions of particular models and uses means-ends analysis to guide it to the desired goal-satisfying model. It later served as the planning system onboard the 1970s Shakey robot.

Fikes , Richard and Nils Nilsson (1972). Learning and Executing generalized robot plans (Technical Note 70. AI Center, SRI International, 333 Ravenswood Ave, Menlo Park, CA 94025.) The first paper to introduce Shakey the Robot.

H. Gelernter, J. R. Hansen, and D. W. of the Loveland (1960). EMPIRICAL EXPLORATIONS OF THE GEOMETRY-THEOREM PROVING MACHINE. Proceedings of the Western Joint Computer Conference, 1960, 17:143-147. (Reprinted in Computers and Thought.)

Green, Bert F. Jr., Alice K. Wolf, Carol Chomsky, and Kenneth Laughery (1961). BASEBALL: AN AUTOMATIC QUESTION ANSWERER. Proc. Western Joint Computer Conference 1961, 19: 219-24. (Reprinted in Computers and Thought.)

Knuth, Donald E. (1965). Semantics for context-free language -- Free abstract available. (Mathematical Systems Theory, 2(2), 127-145).

Licklider, J.C.R. (1960). Man-Computer Symbiosis "Written in 1960, this essay foresaw the growing dependence upon computers for more and more intelligent functions, and an age of human/computer interdependence in which the distinction between the two becomes increasingly blurred. Originally published in IEEE Transactions on Human Factors in Electronics, volume HFE-1, pages 411, March 1960."

• also see: Waldrop, M. Mitchell. 2001. The Origins of Personal Computing. Scientific American 285 (6): 84-91.

Lindsay, Robert K. INFERENTIAL MEMORY AS THE BASIS OF MACHINES WHICH UNDERSTAND NATURAL LANGUAGE. (Reprinted in Computers and Thought.)

McCarthy, et al. (1955). A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence. J. McCarthy, M. L. Minsky, N. Rochester, and C.E. Shannon. August 31, 1955. "We propose that a 2 month, 10 man study of artificial intelligence be carried out during the summer of 1956 at Dartmouth College in Hanover, New Hampshire. The study is to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it." And this marks the debut of the term "artificial intelligence."

McCarthy, John (1959). Programs with Common Sense . "This paper will discuss programs to manipulate in a suitable formal language (most likely a part of the predicate calculus) common instrumental statements. The basic program will draw immediate conclusions from a list of premises. These conclusions will be either declarative or imperative sentences. When an imperative sentence is deduced the program takes a corresponding action. These actions may include printing sentences, moving sentences on lists, and reinitiating the basic deduction process on these lists."

McCarthy, John (1960). Recursive Functions of Symbolic Expressions and Their Computation by Machine, Part I. "This was the original paper on LISP. It is copied with minor notational changes from CACM, April 1960. If you want the exact typography, look there. A few typographical changes have been made, but the notation has not been modernized. There are also some new explanatory footnotes. Part II, which never appeared, was to have had some Lisp programs for algebraic computation." - from his list of Papers on Programming Languages

Miller, George A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81-97. [A classic in memory research and one of the earliest contributions to the "cognitive revolution."]

Minsky, Marvin (1961). STEPS TOWARD ARTIFICIAL INTELLIGENCE. Proceedings of the Institute of Radio Engineers, January, 1961, 49:8-30. What Is Artificial Intelligence? Even though we don't yet understand how brains perform many mental skills, we can still work toward making machines that do the same or similar things. 'Artificial intelligence' is simply the name we give to that research. But as I already pointed out, this means that the focus of that research will keep changing, since as soon as we think we understand one mystery, we have to move on to the next." (Reprinted in Computers and Thought.)

Minsky, Marvin L. (1968). Matter, Mind and Models Published in Semantic Information Processing, MIT Press, 1968. (This version has some editorial simplifications made in 1995)

Minsky, Marvin (1974). A Framework for Representing Knowledge. MIT-AI Laboratory Memo 306, June, 1974. [Reprinted in The Psychology of Computer Vision, P. Winston (Ed.), McGraw-Hill, 1975. Shorter versions in J. Haugeland, Ed., Mind Design, MIT Press, 1981, and in Cognitive Science, Collins, Allan and Edward E. Smith (eds.) Morgan-Kaufmann, 1992 ISBN 55860-013-2] . "Here is the essence of the theory: When one encounters a new situation ... one selects from memory a structure called a Frame. This is a remembered framework to be adapted to fit reality by changing details as necessary."

Newell, A. Heuristic Programming: Ill Structured Problems (1969), in Arnofsky, Julius S,, ed., Progress in Operations Pesearch III. New York: Wiley & Sons. Downloadable PDF. Important paper laying out the distinction between well-structured problems for which formal solution exist and ill-structured problems for which formal (logical & mathematical) methods do not provide precise answers because the problem definition itself is not precisely formed.

Newell, A. and J.C. Shaw (1957). Programming the Logic Theory Machine. Proceedings of the Western Joint Computer Conference, 1957. Downloadable PDF. Additional description of the first running AI program, the Logic Theorist (LT), and the essential programming considerations involved. "ind ways of achieving extreme flexibility. It was developed in connection with a particular substantive problem-proving theorems in symbolic logic-which requires great flexibility in the memory structure, and powerful ways of expressing information processes. The language achieved its purpose: we have a running program for LT which has allowed us to explore its behavior empirically with a number of variations."

Newell, A. , J. C. Shaw, and H. Simon (1957). EMPIRICAL EXPLORATIONS WITH THE LOGIC THEORY MACHINE: A CASE STUDY IN HEURISTICS. Proceedings of the Western Joint Computer Conference, 1957, 15:218--239. A description of the first running AI program, the Logic Theorist (LT). (Reprinted in Computers and Thought.)

A. Newell, J. C. Shaw, and H. Simon (1958) CHESS-PLAYING PROGRAMS AND THE PROBLEM OF COMPLEXITY October, 1958, 2:320-335. (Reprinted in Computers and Thought.)

Newell, Allen and H.A. Simon (1961). GPS, A PROGRAM THAT SIMULATES HUMAN THOUGHT . Lerenden Automaton, Munich: R. Oldenberg KG. (Reprinted in Computers and Thought.)

Samuel , A. L. (1959). Some studies in machine learning using the game of checkers. IBM Journal of Research and Development, July, 1959, 3:211-229. Description of Samuel's tour-de-force checker-playing program that improved its performance through learning.

Selfridge, Oliver G. (1959). Pandemonium: A Paradigm for Learning. Mechanization of Thought Processes, Proc. Teddington Conference, November 1958.

Selfridge Oliver G. , and Ulric Neisser (1960). PATTERN RECOGNITION BY MACHINE. Scientific American (August, 1960) 203: 60-68. (Reprinted in Computers and Thought.)

Simon, H.A. (1961). Modeling Human Mental Processes Proceedings of the Western Joint Computer Conference, 1961. Downloadable PDF.

Slagle, James (1959). A HEURISTIC PROGRAM THAT SOLVES SYMBOLIC INTEGRATION PROBLEMS IN FRESHMAN CALCULUS. One of the very first PhD dissertations on AI. A clear demonstration of the power of heuristic methods to achieve human-level performance in a task domain that is well recognized to require intelligence.

Uhr, Leonard and Charles Vossler (1961). A PATTERN-RECOGNITION PROGRAM THAT GENERATES, EVALUATES, AND ADJUSTS ITS OWN OPERATORS. Proc. Western Joint Computer Conference 1961, 19: 555-570. (Reprinted in Computers and Thought.)

Weizenbaum, J. (1965). ELIZA--A Computer Program for the Study of Natural Language Communication Between Man and Machine. Communications of the ACM, 9 (1): 36-45. A pioneering work in natural language understanding and generation that highlighted the differences between pattern matching and understanding. The ACM makes a .pdf file available online for a fee.

1955 Western Joint Computer Conference Table of Contents. Downloadable versions of several important early papers by Miller, Newell, Selfridge, Ware, and others.

Classic Papers 1975 and Beyond

Agre, Philip E. and David Chapman (1987). Pengi: An Implementation of a Theory of Activity. "AI has generally interpreted the organized nature of everyday activity in terms of plan-following. Nobody could doubt that people often make and follow plans. But the complexity, uncertainty, and immediacy of the real world require a central role for moment-to-moment improvisation." AAAI "Classic Paper" Award in 2006 for the paper's contribution to the field (and resurgence) of reactive planning.

Allen, John (1984). Towards a general theory of action and time -- free abstract available. (Artificial Intelligence 23: 123-154).

Berliner, Hans (1980). The B* Tree Search Algorithm. A Best-First Proof Procedure -- Free abstract available. (Artificial Intelligence 12 (1): 23–40) . An accomplished chess player, Berliner created BKG, a Backgammon evaluation program that later defeated a world champion in the game.

Buchanan, B. G. , D. H. Smith, W. C. White, R. J. Gritter, E. A. Feigenbaum, J. Lederberg, Carl Djerassi (1976). Applications of artificial intelligence for chemical inference. 22: Automatic rule formation in mass spectrometry by means of the meta-DENDRAL program. J. Am. Chem. Soc., 1976, 98 (20), pp 6168–6178.. First published discovery of a scientific principle by a computer program in the refereed scientific literature. <PDF of first page free, full article available free to subscribers and libraries only.>

Brooks, Rodney (1986) . A robust, layered, control system for a mobile robot (IEEE Journal of Robotics and Automation, Vol. 2, No. 1, March 1986, pp. 14–23; also MIT AI Memo 864).

Canny, John (1983). A Variational Approach to Edge Detection. "The problem of detecting intensity changes in images is canonical in vision. Edge detection operators are typically designed to optimally estimate first or second derivative over some (usually small) support. Other criteria such as output signal to noise ratio or bandwidth have also been argued for. This paper describes an attempt to formulate set of edge detection criteria that capture as directly as possible the desirable properties of the detector." AAAI "Classic Paper" Award in 2002 in recognition of the wide use of the Canny Edge Detector introduced in this paper as well as seminal contributions in the areas of robotics and machine perception.

Cheeseman, Peter ,Matthew Self, Jim Kelly, Will Taylor, and Don Freeman (1988). Bayesian Classification. "This paper describes a Bayesian technique for unsupervised classification of data and its computer implementation, AutoClass." AAAI "Classic Paper" Award in 2007 in recognition of its foundational theoretical and practical contributions to machine learning.

Erman, Lee D. and Frederick Hayes-Roth, Victor R. Lesser, D. Raj Reddy (1980). The Hearsay-II Speech-Understanding System: Integrating Knowledge to Resolve Uncertainty. ACM Computing Surveys 12(2): 213 - 253. "The Hearsay-II speech-understanding system ... recognizes connected speech in a 1000-word vocabulary with correct interpretations for 90 percent of test sentences. Its basic methodology involves the application of symbolic reasoning as an aid to signal processing. A marriage of general artificial intelligence techniques with special acoustic and linguistic knowledge was needed to accomplish satisfactory speech-understanding performance." <Available for free to subscribers only.>

Hanks, Steve and Drew McDermott (1986). Default Reasoning, Nonmonotonic Logics, and the Frame Problem. "Nonmonotonic formal systems have been proposed as an extension to classical first-order logic that will capture the process of human "default reasoning" or "plausible inference" through their inference mechanisms just as modus ponens provides a model for deductive reasoning. But although the technical properties of these logics have been studied in detail and many examples of human default reasoning have been identified, for the most part these logics have not actually been applied to practical problems to see whether they produce the expected results....Upon analyzing the failure we find that the nonmonotonic logics we considered are inherently incapable of representing this kind of default reasoning." AAAI "Classic Paper" Awardin 2005 for their work contradicting the use of default reasoning on the frame problem.

Haussler, David (1986). Quantifying the Inductive Bias in Concept Learning. This paper shows "that the notion of bias in inductive concept learning can be quantified in a way that directly relates to learning performance, and that this quantitative theory of bias can provide guidance in the design of effective learning algorithms." AAAI "Classic Paper" Award in 2005 in recognition of Haussler's work in machine learning algorithms as well as the introduction of PAC learning from a theoretical standpoint.

Lansky, Amy and Michael P. Georgeff (1987). Reactive Reasoning and Planning. "In this paper a robot is equipped with reasoning and planning capabilities of that allow it to need only be partly elaborated before it decides to act. This allows the robot to avoid overly strong expectations about the environment, overly constrained plans of action, and other forms of overcommitment common to previous planners." AAAI "Classic Paper" Award in 2006 for the paper's contribution to the long-term resurgence of integrative planning-execution systems.

Lenat, Doug, Mayank Prakash, and Mary Shepherd. CYC: Using common sense knowledge to overcome brittleness and knowledge acquisition bottlenecks. AI Magazine, 6(4):65-85, 1985. An early paper setting forth the long-term goals of the CYC project.

Levesque, Hector (1984). A Logic of Implicit and Explicit Belief. "In this paper, we point out deficiencies in current semantic treatments of knowledge and belief (including recent syntactic approaches) and suggest a new analysis in the form of a logic that avoids these shortcomings and is also more viable computationally." AAAI "Classic Paper" Awardin 2004 for his work in knowledge representation, notably his contradiction of current practices.

McDermott , John (1980). R1: An Expert in the Computer Systems Domain. "R1 is a rule-based system that has much in common with other domain-specific systems that have been developed over the past several years. It differs from these systems primarily in its use of Match rather than Generate-and-Test as its central problem solving method; rather than exploring several hypotheses until an acceptable one is found, it exploits its knowledge of its task domain to generate a single acceptable solution." AAAI "Classic Paper" Award in 1999.

Minton, Steven ,Mark D. Johnston, Andrew B. Philips, and Philip Laird (1990). Solving Large-Scale Constraint Satisfaction and Scheduling Problems Using a Heuristic Repair Method. "This paper describes a simple heuristic method for solving large-scale constraint satisfaction and scheduling problems. Given an initial assignment for the variables in a problem, the method operates by searching though the space of possible repairs. The search is guided by an ordering heuristic, the min-conflicts heuristic, that attempts to minimize the number of constraint violations after each step... A theoretical analysis is presented to explain why the method works so well on certain types of problems and to predict when it is likely to be most effective." AAAI "Classic Paper" Award in 2008 for a seminal contribution to stochastic local search for constraint satisfaction and its broad influence on local search algorithms and applications in artificial intelligence.

Muggleton, S.H. (1991). Inductive Logic Programming. New Generation Computing, 8(4):295-318, 1991. An early exposition of inductive logic programming.

Newell, Allen (1980). Newell: The Knowledge Level. By Allen Newell. AAAI Presidential Address, 19 August 1980. AI Magazine 2(2): Summer 1981, 1-20, 33. A classic article describing the differences in viewing computer programs at the symbol level or the knowledge level.

Newell, Allen (1992). Newell: Fairy Tales. By Allen Newell. AI Magazine 13(4): Winter 1992, 46-4. In this reprint of Allen Newell's classic essay, Newell argues not only that fairy tales are for all of us, but that, even more, they have a close connection to technology.

Newell, Allen & Herb Simon (1975). Computer Science as Empirical Inquiry. Turing Award Lecture, Communications of the ACM 19 (3), (March 1976): p. 113–126. " Computer science is an empirical discipline. ... Each new machine that is built is an experiment. Actually constructing the machine poses a question to nature; and we listen for the answer by observing the machine in operation and analyzing it .... Each new program that is built is an experiment...." This paper contains the Physical Symbol System Hypothesis:

 "A physical symbol system has the necessary and sufficient means for intelligent action."

Nii, Penny. Blackboard systems. AI Magazine, 7(2), 1986. "The first blackboard system was the HEARSAY-II speech understanding system (Erman et al.,1980) that evolved between 1971 and 1976. Subsequently, many systems have been built that have similar system organization and run-time behavior. The objectives of this article are (1) to define what is meant by "blackboard systems" and (2) to show the richness and diversity of blackboard system designs. The article begins with a discussion of the underlying concept behind all blackboard systems, the blackboard model of problem solving. In order to bridge the gap between a model and working systems, the blackboard framework, an extension of the basic blackboard model is introduced, including a detailed description of the model's components and their behavior. A model does not come into existence on its own, and is usually an abstraction of many examples. In Section 2 the history of ideas is traced, and the designs of some application systems that helped shape the blackboard model are detailed. "

Pearl , Judea (1982). Reverend Bayes on Inference Engines: A Distributed Hierarchical Approach. "This paper presents generalizations of Bayes likelihood-ratio updating rule which facilitate an asynchronous propagation of the impacts of new beliefs and/or new evidence in hierarchically organized inference structures with multi-hypotheses variables." AAAI "Classic Paper" Award in 2000 for revolutionizing uncertain reasoning through the introduction of efficient Bayesian inference methods. See also Judea Pearl (1985) Bayesian Networks: A Model of Self-Activated Memory for Evidential Reasoning (UCLA Technical Report CSD-850017). (Proceedings of the 7th Conference of the Cognitive Science Society, University of California, Irvine, CA. pp. 329–334.)

Quinlan , John Ross (1979). Discovering Rules from Large collection of examples (Expert Systems in the Microelectronic Age, Edinburgh University Press, Edinburgh, Scotland).

Reddy , Raj (1995). Reddy: To Dream The Possible Dream. Raj Reddy's Turing Award Lecture presented at the ACM CS Conference, March 1, 1995. "This essay is collection of retrospective and prospective remarks on the role of AI within CS and in society. It includes comments on questions such as: Can artificial intelligence equal human intelligence? Isn't AI just a special class of algorithms? Isn't AI just software? Why should society support AI and CS research? What next for AI? And so on. The main theme is that AI continues to be a possible dream worthy of dreaming."

Selfridge, Oliver G. (1993). The Gardens of Learning: A Vision for AI. AI Magazine 14(2): Summer 1993, 36-48. "I have watched AI since its beginnings ... In 1943, I was an undergraduate at the Massachusetts Institute of Technology (MIT) and met a man whom I was soon to be a roommate with. He was but three years older than I, and he was writing what I deem to be the first directed and solid piece of work in AI. His name was Walter Pitts, and he had teamed up with a neurophysiologist named Warren McCulloch, who was busy finding out how neurons worked (McCulloch and Pitts 1943)."

Related Resources

• A few online classic books in AI are collected on this site.
• Recent Papers that have been most frequently cited

• Histories of Specialized Topics Several lists of papers of historic interest from the histories of various specialty areas including Cognitive Science, Decision Making, and Technology]]

Criteria for Inclusion

Classic papers are those that have had a substantial impact on the science or practice of AI. We include them as we find copies online, or as we obtain permissions to copy them and put them on the AAAI server. All of the AAAI award winners are included plus papers 15 years old and older that have been cited several hundred times. Suggestions for additional papers are welcome.

Work in Progress -- Looking for Online Versions

About the AAAI "Classic Paper" Award

The AAAI Classic Paper award honors the author(s) of paper(s) deemed most influential, chosen from a specific conference year. Papers were judged on the basis of impact, for example those that started a new research (sub)area, led to important applications, answered a long-standing question/issue or clarified what had been murky, made a major advance that figures in the history of the subarea, has been picked up as important and used by other areas within (or outside of) AI and/or has been very heavily cited.