Skip to main content.

About

This course explores the use of "declarative languages" for programming. Such programming languages enable specifying the characteristics of a solution to a problem, rather than specifying the steps that have to be followed to come to this solution. The system itself will find the solution using a general-purpose problem solving procedure. In most declarative programming languages, the problem is specified by means of assertions in a logic. The problem solving strategy is a proof procedure which can be queried for answers that follow from the assertions. We will explore the theoretical underpinnings of the declarative programming paradigm, the diveristy among the programming languages that different logics and proof procedures gives rise to and their applications in the domain of artificial intelligence for which they are particulary well-suited.

The official course description can be found here and here.

The exam consists of an oral test with written preparation about the entire course (theory and exercises) and an oral defense of an individually completed programming project. The end result is calculated as the average of the results on both parts. If one of both results is 7 or less, however, the end result cannot exceed 7.

When and Where

The theory is lectured every Wednesday from 13:00 till 15:00 in room D.2.19 (in room F.4.104 on the 1st of December).

The exercise sessions are organized after the lectures (from 15:00 till 17:00) of October 6th, 13th, 20th, 27th and November 3rd in room E.1.2.

Course Material

Peter Flach, "Simply Logical: Intelligent Reasoning by Example", J. Wiley & Sons, 1994 , ISBN 0471 94152 2
This book is currently out of print, but available for free on-line.

The material for the individual lectures can be found below (or here as one large [PDF] or one with 4 slides per page [PDF]).

Introduction
Completed on 29/09/2010
Theoretical Backgrounds
Completed on 13/10/2010
Logic programming and Prolog
Completed on 3/11/2010
  • Slides: SLD-resolution refutation, pruning the search, negation as failure [PDF], tail recursion optimization, arithmetic, difference lists, second-order predicates, failure-driven loops [PDF], higher-order programming, inspecting terms, extending Prolog (through term expansion, custom operators and meta-interpretation), a methodology for logic programming [PDF], revisiting the Eliza classic in Prolog [PDF]
  • Further reading: Chapter 3 of Simply Logical, The Logic Programming Paradigm and Prolog is highly recommended as it summarizes the above from an alternative point of view, A couple of meta-interpreters in Prolog
Blind and informed search through state spaces and proving as a search process
Completed on 10/11/2010
Natural language processing using definite clause grammars
Completed on 17/11/2010
Efficiency issues, best practices and coding guidelines
Group discussion on 1/12/2010
Reasoning with incomplete information
Completed on 8/12/2010
Interesting loose ends (will not be asked on the exam)
Completed on 15/12/2010

Additional Material