"Computer programming is the process of constructing executable code from fragmentary information. ... When computer programming is done by a machine, the process is called automatic programming. AI researchers are interested in studying automatic programming for two reasons: First, it would be highly useful to have a powerful automatic programming systems that could receive casual and imprecise specifications for a desired target program and then correctly generate that program; second, automatic programming is widely believed to be a necessary component of any intelligent system and is therefore a topic for fundamental research in its own right."
– excerpt from Biermann, A. 1992. Automatic Programming. In Encyclopedia of Artificial Intelligence. 2nd edition, Stuart C. Shapiro, editor, 18 - 35. New York: John Wiley & Sons.
AR&A techniques have been used to solve a variety of tasks, including automatic programming, constraint satisfaction, design, diagnosis, machine learning, search, planning, reasoning, game playing, scheduling, and theorem proving. The primary purpose of AR&A techniques in such settings is to overcome computational intractability. In addition, AR&A techniques are useful for accelerating learning and summarizing sets of solutions. The Fifth Symposium on Abstraction, Reformulation, and Approximation (SARA-2002) was held from 2 to 4 August 2002, directly after the Eighteenth National Conference on Artificial Intelligence (AAAI-2002). It was chaired by Sven Koenig from the Georgia Institute of Technology and Robert Holte from the University of Alberta (Canada) and held at Kananaskis Mountain Lodge, Kananaskis Village, Alberta (Canada) between Calgary and Banff in the Rocky Mountains.
Most work in automatic programming has focused primarily on the roles of deduction and programming knowledge However, the role played by knowledge of the task domain seems to be at least as important, both for the usability of an automatic programming system and for the feasibility of building one which works on nontrivial problems This perspective has evolved during the course of a variety of studies over the last several years, including detailed examination of existing software for a particular domain (quantitative interpretation of oil well logs) and the implementation of an experimental automatic programming system for that domain The importance of domain knowledge has two important implications: a primary goal of automatic programming research should be to characterize the programming process for specific domains; and a crucial issue to be addressed in these characterizations is the interaction of domain and programming knowledge during program synthesis Used by permission of the International Joint Conferences on Artificial Intelligence; copies of the Proceedings are available from William Kaufmann, Inc, 95 First St., Los Altos, CA 94022 USA. For example, the work of Green (1969) and Waldinger and Lee (1969) in the late 1960s was concerned with the use of a theorem-prover to produce programs. This deductive paradigm continues to be the basis for much research in automatic programming (e.g., Manna & Waldinger 1980, Smith 1983). In the mid 1970's, work on the PSI project (Barstow 1979, Green 1977, Kant 1981) and on the Programmer's Apprentice (Rich 1981) was fundamentally concerned with the codification of knowledge about programming techniques and the use of that knowledge in program synthesis and analysis Work within the knowledge-based paradigm is also continuing (e.g., Barstow 1982, Waters 1981). This article is concerned with the role played by knowledge of the task domain, a role which seems to be at least as important.
The course consists of lectures for the first two-thirds of the semester. Homework problems and programming assignments illustrate the lecture material. The programs are not long; the intent is to gain some exposure to several kinds of programming systems. The latter part of the semester covers readings in the research literature; students are expected to present one or two papers to the class. Many of the world's best researchers in automatic programming are in Austin: Jim Browne, Don Batory, Elaine Kant, Ira Baxter, Ted Biggerstaff; they will be invited to present guest lectures to describe their work.
Our goal is automatic generation of computer programs from specifications that are much smaller and easier to write than ordinary programs. Generation of geometric programs specified by diagrams 2011 Yulin Li and Gordon S. Novak, Jr., In Proceedings of the 10th ACM international conference on Generative programming and component engineering, pp. Computer aided software design via inference and constraint propagation 2009 Gordon Novak, Integrated Computer-Aided Engineering, Vol. 16, 3 (2009), pp. Knowledge Based Programming Using Abstract Data Types 1983 Gordon Novak, In Proc.
Our approach to automatic programming is based on reuse of generic algorithms through views. A generic algorithm performs some task, such as sorting a linked list of records, based on abstract descriptions of the data on which the program operates. A view describes how actual application data corresponds to the abstract data as used in the generic algorithm. Given a view, a generic algorithm can be specialized by a compilation process to produce a version of the algorithm that performs the algorithm directly on the application data.
Professor Martin, whose main interest was in the practical application of artificial intelligence, was associated with the Laboratory for Computer Science and the Artificial Intelligence Laboratory at MIT. His work involved computer programs that embody various forms of expertise---mathematical, medical, management or linguistic---and the application of this expertise to practical ends. His system for symbolic mathematics is now used around the country by a large community of scientists. Professor Martin's interest in mangagement led him to the development of automatic programming techniques that are widely used.
An ability to write functionally correct programs -- those that pass test cases? A seasoned interviewer would tell you that there is much more to writing code than passing test cases! For starters, we really care for how well a candidate understands the problem and approaches a solution than being able to write functionally correct code. Machine learning has helped solved many grading challenges -- spoken english, essay grading, program grading and math problem grading to cite a few examples.
John McCarthy, born at Boston, Mass. in 1927, received his B.S. degree in mathematics at the California Institute of Technology in 1948, and his Ph.D. also in mathematics at Princeton University in 1951. He is at present Assistant Professor of Communication Sciences at the Massachusetts Institute of Technology. His present interests are in the artificial intelligence problem, automatic programming and mathematical logic. He is co-editor with Dr. C. E. Shannon of "Automatic Studies". However, certain elementary verbal reasoning processes so simple that they can be carried out by any non--feeble--minded human have yet to be simulated by machine programs.
Douglas B. Lenati December, l99,9 Abstract DNA may be regarded as a "program" for constructing and maintaining an organism. The field of Automatic Programming studies computer programs which synthesize new and different programs, or which modify and improve themselves. When DNA molecules do this, we call it Evolution. Biological research has to date identified several mechanisms which change DNA (substitution, insertion, deletion, translocation, inversion, recombination, segregation, transposition, etc.) Current theories assume the basic process of evolution to be Early automatic programming systems were also built to work via this same "Random Generate and Test" process. But that mechanism failed, and we now recognize the reasons for that failure and the prescription for success.
Department of Computer Science, Duke University, Durham, North Carolina 27706, U.S.A. (Received 4 March 1985) Ten methodologies for automatic program construction are presented, discussed and compared. Some of the techniques generate code from formal input--output specifications while others work from examples of the target behaviour or from natural language input. Automatic computer programming or automatic programming occurs whenever a machine aids in this process. The amount of automatic programming that is occurring is a variable quantity that depends on how much aid the human is given. There are a number of dimensions on which the level of help can be measured including the level of the language used by the human, the amount of informality allowed, the degree to which the system is told what to do rather than how to do it, and the efficiency of the resulting code.