Derivación deductiva de programas funcionales con patrones

Una de las dificultades principales en el desarrollo de software es la ausencia de un marco conceptual adecuado para su estudio. Una propuesta la constituye el modelo transformativo, que entiende el desarrollo de software como un proceso iterativo de transformación de especificaciones: se parte de una especificación inicial que va transformándose sucesivamente hasta obtener una especificación final que se toma como programa. Este modelo básico puede llevarse a la práctica de varias maneras. En concreto, la aproximación deductiva toma una sentencia lógica como especificación inicial y su proceso transformador consiste en la demostración de la sentencia; como producto secundario de la demostración se deriva un programa que satisface la especificación inicial. La tesis desarrolla un método deductivo para la derivación de programas funcionales con patrones, escritos en un lenguaje similar a Hope. El método utiliza una lógica multigénero, cuya relación con el lenguaje de programación es estudiada. También se identifican los esquemas de demostración necesarios para la derivación de funciones con patrones, basados en la demostración independiente de varias subsentencias. Cada subsentencia proporciona una subespecificación de una ecuación del futuro programa a derivar. Nuestro método deductivo está inspirado en uno previo de Zohar Manna y Richard Waldinger, conocido como el cuadro deductivo, que deriva programas en un lenguaje similar a Lisp. El nuevo método es una modificación del cuadro de estos autores, que incorpora géneros y permite demostrar una especificación mediante varios cuadros. Cada cuadro demuestra una subespecificación y por tanto deriva una ecuación del programa. Se prevén mecanismos para que los programas derivados puedan contener definiciones locales con patrones y variables anónimas y sinónimas y para que las funciones auxiliares derivadas no usen variables de las funciones principales. La tesis se completa con varios ejemplos de aplicación, un mecanismo que independentiza el método del lenguaje de programación y un prototipo de entorno interactivo de derivación deductiva. Categorías y descriptores de materia CR D.l.l [Técnicas de programación]: Programación funcional; D.2.10 [Ingeniería de software]: Diseño - métodos; F.3.1 [Lógica y significado de los programas]: Especificación, verificación y razonamiento sobre programas - lógica de programas; F.3.3 [Lógica y significado de los programas]: Estudios de construcciones de programas - construcciones funcionales; esquemas de programa y de recursion; 1.2.2 [Inteligencia artificial]: Programación automática - síntesis de programas; 1.2.3 [Inteligencia artificial]: Deducción y demostración de teoremas]: extracción de respuesta/razón; inducción matemática. Términos generales Programación funcional, síntesis de programas, demostración de teoremas. Otras palabras claves y expresiones Funciones con patrones, cuadro deductivo, especificación parcial, inducción estructural, teorema de descomposición.---ABSTRACT---One of the main difficulties in software development is the lack of an adequate conceptual framework of study. The transformational model is one such proposal that conceives software development as an iterative process of specifications transformation: an initial specification is developed and successively transformed until a final specification is obtained and taken as a program. This basic model can be implemented in several ways. The deductive approach takes a logical sentence as the initial specification and its proof constitutes the transformational process; as a byproduct of the proof, a program which satisfies the initial specification is derived. In the thesis, a deductive method for the derivation of Hope-like functional programs with patterns is developed. The method uses a many-sorted logic, whose relation to the programming language is studied. Also the proof schemes necessary for the derivation of functional programs with patterns, based on the independent proof of several subsentences, are identified. Each subsentence provides a subspecification of one equation of the future program to be derived. Our deductive method is inspired on a previous one by Zohar Manna and Richard Waldinger, known as the deductive tableau, which derives Lisp-like programs. The new method incorporates sorts in the tableau and allows to prove a sentence with several tableaux. Each tableau proves a subspecification and therefore derives an equation of the program. Mechanisms are included to allow the derived programs to contain local definitions with patterns and anonymous and synonymous variables; also, the derived auxiliary functions cannot reference parameters of their main functions. The thesis is completed with several application examples, i mechanism to make the method independent from the programming language and an interactive environment prototype for deductive derivation. CR categories and subject descriptors D.l.l [Programming techniques]: Functional programming; D.2.10 [Software engineering]: Design - methodologies; F.3.1 [Logics and meanings of programa]: Specifying and verifying and reasoning about programs - logics of programs; F.3.3 [Logics and meanings of programs]: Studies of program constructs - functional constructs; program and recursion schemes; 1.2.2 [Artificial intelligence]: Automatic programming - program synthesis; 1.2.3 [Artificial intelligence]: Deduction and theorem proving - answer/reason extraction; mathematical induction. General tenas Functional programming, program synthesis, theorem proving. Additional key words and phrases Functions with patterns, deductive tableau, structural induction, partial specification, descomposition theorem.

A Comparison of two Proof Critics: Power vs. Robustness

Proof critics are a technology from the proof planning paradigm. They examine failed proof attempts in order to extract information which can be used to generate a patch which will allow the proof to go through. We consider the proof of the $quot;whisky problem$quot;, a challenge problem from the domain of temporal logic. The proof requires a generalisation of the original conjecture and we examine two proof critics which can be used to create this generalisation. Using these critics we believe we have produced the first automatic proofs of this challenge problem. We use this example to motivate a comparison of the two critics and propose that there is a place for specialist critics as well as powerful general critics. In particular we advocate the development of critics that do not use meta-variables.

Privileged Words and Sturmian Words

This dissertation has two almost unrelated themes: privileged words and Sturmian words. Privileged words are a new class of words introduced recently. A word is privileged if it is a complete first return to a shorter privileged word, the shortest privileged words being letters and the empty word. Here we give and prove almost all results on privileged words known to date. On the other hand, the study of Sturmian words is a well-established topic in combinatorics on words. In this dissertation, we focus on questions concerning repetitions in Sturmian words, reproving old results and giving new ones, and on establishing completely new research directions. The study of privileged words presented in this dissertation aims to derive their basic properties and to answer basic questions regarding them. We explore a connection between privileged words and palindromes and seek out answers to questions on context-freeness, computability, and enumeration. It turns out that the language of privileged words is not context-free, but privileged words are recognizable by a linear-time algorithm. A lower bound on the number of binary privileged words of given length is proven. The main interest, however, lies in the privileged complexity functions of the Thue-Morse word and Sturmian words. We derive recurrences for computing the privileged complexity function of the Thue-Morse word, and we prove that Sturmian words are characterized by their privileged complexity function. As a slightly separate topic, we give an overview of a certain method of automated theorem-proving and show how it can be applied to study privileged factors of automatic words. The second part of this dissertation is devoted to Sturmian words. We extensively exploit the interpretation of Sturmian words as irrational rotation words. The essential tools are continued fractions and elementary, but powerful, results of Diophantine approximation theory. With these tools at our disposal, we reprove old results on powers occurring in Sturmian words with emphasis on the fractional index of a Sturmian word. Further, we consider abelian powers and abelian repetitions and characterize the maximum exponents of abelian powers with given period occurring in a Sturmian word in terms of the continued fraction expansion of its slope. We define the notion of abelian critical exponent for Sturmian words and explore its connection to the Lagrange spectrum of irrational numbers. The results obtained are often specialized for the Fibonacci word; for instance, we show that the minimum abelian period of a factor of the Fibonacci word is a Fibonacci number. In addition, we propose a completely new research topic: the square root map. We prove that the square root map preserves the language of any Sturmian word. Moreover, we construct a family of non-Sturmian optimal squareful words whose language the square root map also preserves.This construction yields examples of aperiodic infinite words whose square roots are periodic.

Reasoning from Incomplete Knowledge in a Procedural Deduction System

One very useful idea in AI research has been the notion of an explicit model of a problem situation. Procedural deduction languages, such as PLANNER, have been valuable tools for building these models. But PLANNER and its relatives are very limited in their ability to describe situations which are only partially specified. This thesis explores methods of increasing the ability of procedural deduction systems to deal with incomplete knowledge. The thesis examines in detail, problems involving negation, implication, disjunction, quantification, and equality. Control structure issues and the problem of modelling change under incomplete knowledge are also considered. Extensive comparisons are also made with systems for mechanica theorem proving.

Introduction to logical information systems

Ridoux, O. and Ferr?, S. (2004) Introduction to logical information systems. Information Processing & Management, 40 (3), 383-419. Elsevier

Building Responsive Systems from Physically-correct Specifications

Predictability - the ability to foretell that an implementation will not violate a set of specified reliability and timeliness requirements - is a crucial, highly desirable property of responsive embedded systems. This paper overviews a development methodology for responsive systems, which enhances predictability by eliminating potential hazards resulting from physically-unsound specifications. The backbone of our methodology is the Time-constrained Reactive Automaton (TRA) formalism, which adopts a fundamental notion of space and time that restricts expressiveness in a way that allows the specification of only reactive, spontaneous, and causal computation. Using the TRA model, unrealistic systems - possessing properties such as clairvoyance, caprice, in finite capacity, or perfect timing - cannot even be specified. We argue that this "ounce of prevention" at the specification level is likely to spare a lot of time and energy in the development cycle of responsive systems - not to mention the elimination of potential hazards that would have gone, otherwise, unnoticed. The TRA model is presented to system developers through the CLEOPATRA programming language. CLEOPATRA features a C-like imperative syntax for the description of computation, which makes it easier to incorporate in applications already using C. It is event-driven, and thus appropriate for embedded process control applications. It is object-oriented and compositional, thus advocating modularity and reusability. CLEOPATRA is semantically sound; its objects can be transformed, mechanically and unambiguously, into formal TRA automata for verification purposes, which can be pursued using model-checking or theorem proving techniques. Since 1989, an ancestor of CLEOPATRA has been in use as a specification and simulation language for embedded time-critical robotic processes.

OS Support for Portable Bulk Synchronous Parallel Programs

Predictability -- the ability to foretell that an implementation will not violate a set of specified reliability and timeliness requirements -- is a crucial, highly desirable property of responsive embedded systems. This paper overviews a development methodology for responsive systems, which enhances predictability by eliminating potential hazards resulting from physically-unsound specifications. The backbone of our methodology is the Time-constrained Reactive Automaton (TRA) formalism, which adopts a fundamental notion of space and time that restricts expressiveness in a way that allows the specification of only reactive, spontaneous, and causal computation. Using the TRA model, unrealistic systems – possessing properties such as clairvoyance, caprice, infinite capacity, or perfect timing -- cannot even be specified. We argue that this "ounce of prevention" at the specification level is likely to spare a lot of time and energy in the development cycle of responsive systems -- not to mention the elimination of potential hazards that would have gone, otherwise, unnoticed. The TRA model is presented to system developers through the Cleopatra programming language. Cleopatra features a C-like imperative syntax for the description of computation, which makes it easier to incorporate in applications already using C. It is event-driven, and thus appropriate for embedded process control applications. It is object-oriented and compositional, thus advocating modularity and reusability. Cleopatra is semantically sound; its objects can be transformed, mechanically and unambiguously, into formal TRA automata for verification purposes, which can be pursued using model-checking or theorem proving techniques. Since 1989, an ancestor of Cleopatra has been in use as a specification and simulation language for embedded time-critical robotic processes.

Mining inconsistent secure messages toward analyzing security protocols

Traditional approaches such as theorem proving and model checking have been successfully used to analyze security protocols. Ideally, they assume the data communication is reliable and require the user to predetermine authentication goals. However, missing and inconsistent data have been greatly ignored, and the increasingly complicated security protocol makes it difficult to predefine such goals. This paper presents a novel approach to analyze security protocols using association rule mining. It is able to not only validate the reliability of transactions but also discover potential correlations between secure messages. The algorithm and experiment demonstrate that our approaches are useful and promising.

Quasilinear dirichlet problems in RN with critical growth

In this study, a given quasilinear problem is solved using variational methods. In particular, the existence of nontrivial solutions for GP is examined using minimax methods. The main theorem on the existence of a nontrivial solution for GP is detailed.

Neural network detection of grinding burn from acoustic emission

An artificial neural network (ANN) approach is proposed for the detection of workpiece `burn', the undesirable change in metallurgical properties of the material produced by overly aggressive or otherwise inappropriate grinding. The grinding acoustic emission (AE) signals for 52100 bearing steel were collected and digested to extract feature vectors that appear to be suitable for ANN processing. Two feature vectors are represented: one concerning band power, kurtosis and skew; and the other autoregressive (AR) coefficients. The result (burn or no-burn) of the signals was identified on the basis of hardness and profile tests after grinding. The trained neural network works remarkably well for burn detection. Other signal-processing approaches are also discussed, and among them the constant false-alarm rate (CFAR) power law and the mean-value deviance (MVD) prove useful.

PI Stabilization of a Class of Time Delay Systems

In this paper we use the Hermite-Biehler theorem to establish results for the design of proportional plus integral (PI) controllers for a class of time delay systems. We extend results of the polynomial case to quasipolynomials using the property of interlacing in high frequencies of the class of time delay systems considered. A signature for the quasipolynomials in this class is derived and used in the proposed approach which yields the complete set of the stabilizing PI controllers.

PID stabilization of a class of time delay systems

In this paper we use the Hermite-Biehler theorem to establish results for the design of proportional plus integral plus derivative (PID) controllers concerning a class of time delay systems. Using the property of interlacing at high frequencies of the class of systems considered and linear programming we obtain the set of all stabilizing PID controllers. © 2005 IEEE.