William E. Ball

Formal derivation of rule-based programs

It is shown that a combination of specification and program refinement may be applied to deriving efficient concurrent rule-based programs. Specification refinement is used to generate an initial rule-based program that is refined into a program which is highly concurrent and efficient. This program derivation strategy is divided into two major tasks. The first task relies on specification refinement. Techniques similar to those employed in the derivation of UNITY programs are used to produce a correct rule-based program having a static knowledge base. The second task involves program refinement and is specific to the development of concurrent rule-based programs. It relies heavily on the availability of a computational model, such as Swarm, that has the ability to dynamically restructure the knowledge base. The ways in which a Swarm program can be translated to OPS5 specifically, given some restrictions, while maintaining the correctness criteria are discussed. >

Formal verification of pure production system programs

Reliability, defined as the guarantee that a program satisfies its specifications, is an important aspect of many applications for which rule-based programs are suited. Executing rule-based programs on a series of test cases does not guarantee correct behavior in all possible test cases. To show a program is reliable, it is desirable to construct formal specifications for the program and to prove that it obeys those specifications. This paper presents an assertional approach to the verification of a class of rule-based programs characterized by the absence of conflict resolution. The proof logic needed for verification is already in use by researchers in concurrent programming. The approach involves expressing the program in a language called Swarm, and its specifications as assertions over the Swarm program. Among models that employ rule-based notation, Swarm is the first to have an axiomatic proof logic. A brief review of Swarm and its proof logic is given, along with an illustration of the formal verification method used on a simple rule-based program.

Rapid search for spherical objects in aerial photographs

A methodology is presented for designing detectors which locate specific features in an image. The method is applied to the detection and segmentation of spherical features. A vital part of the detection and segmentation is the use of the gradient angle transform. An analysis of the gradient angle for ideal spheres is presented, with a discussion of how this may be used to locate the boundaries of the sphere. The algorithms used by a program which detects and segments spherical features are then presented. The results of applying the program to images with man-made spherical features are given.<<ETX>>

A multiscale approach for recognizing complex annotations in engineering documents

A novel method for character recognition targeted at complex annotations found in engineering documents is presented. A feasibility study is described in which characters extracted from engineering drawings were recognized without error from a class of 36 distinct alphanumeric patterns by a neural network classifier trained with multiscale representations. An incremental strategy is presented for resolution which relies upon the continuity between hierarchical levels of a novel multiscale decomposition. The authors observed a 16-fold reduction in the amount of information needed to represent each character for recognition. These results suggest high reliability at a reduced cost of representation.<<ETX>>

AUTOMAST: automatic mathematical analysis and symbolic translation

A procedure for numerically solving systems of ordinary differential equations is shown to also generate symbolic solutions. The procedure is based on a finite Taylor series expansion that includes an estimate of the error in the final result. A computer program is described that reads in a system of such equations and then generates the expansions for all of the dependent variables. The expansions are determined symbolically, hence any non-numeric parameters in the original equations are carried automatically into the final expansions. Thus the exact influence of any parameters on the problem solution can be easily displayed.

Correct parallel status assignment for the reason maintenance system

This paper represents a beginning development of a parallel truth maintenance system to interact with a parallel inference engine. We present a solution which performs status assignments in parallel to belief nodes in the Reason Maintenance System (RMS) presented by [Doyle, 1979], [Doyle, 1983]. We examine a previously described algorithm by [Petrie, 1986] which fails to correctly detect termination of the status assignments. Under Petries algorithm, termination may go undetected and in certain circumstances (namely the existence of an unsatisfiable circularity) a false detection may occur. We present an algorithm that corrects these problems.

Toward a fully parallel reason maintenance system

A truth maintenance system (TMS) is an AI system used to monitor consistency of information in a knowledge base. A TMS may be necessary when non-monotonic reasoning is used since incorrect assumptions can lead to contradictory conclusions. The Reason Maintenance System (RMS), a specific TMS first described by Doyle [5],[6], is used along with an inference engine (IE), or problem solver, to maintain a consistent set of beliefs and inferences. We have developed a parallel version of the RMS for correctly assigning IN or OUT states to each believe node [7]. This algorithm uses diffusing computation [4] to assign... Read complete abstract on page 2.

Seeking concurrency in rule-based programming

This paper describes a formal approach for developing concurrent rule-based programs. Specification refinement is used to generate an initial version of the program. Program refinement is then applied to produce a highly concurrent and efficient version of the same program. Techniques for deriving concurrent programs through either specification or program refinement have been described in previous literature. The main contribution of this paper consists of extending the applicability of these techniques to a broad class of rule-based programs. To the best of our knowledge, this is the first time formal derivation is employed in the context of rule-based programming.

BEMAS: User's Manual, 2nd Edition

This paper is a users manual for BEMAS, a BE lief MA intenance System. BEMAS is a menu driven system which provides an easy to use interface between a user and a knowledge base. Given a set of data, and a set of rules, BEMAS will help the user to identify an object by analyzing the properties of that object. Data can be added and deleted at any time, either directly or by deleting beliefs on which the data is dependent. BEMAS maintains the relations and dependencies between data using a dynamic dependency net. BEMAS also has the capability to make inferences using incomplete information while still maintaining knowledge base integrity.

A Survey of Three Applications of Parallelism in AI

Once the BEMAS [13] system was completed and recorded in Common Lisp, research efforts were channeled toward three primary areas. This report will present a briefly review of some research in these areas, which are: parallelizing truth maintenance systems, parallelizing production systems, and parallel search. The area of parallel search has been studied by many over the past years and we will only present current research that has been accomplished. This review represents the beginning research into the development of a parallel inference model.