Given an argumentation framework AF, we introduce a mapping function that constructs a disjunctive logic program P, such that the preferred extensions of AF correspond to the stable models of P, after intersecting each stable model with the relevant atoms. The given mapping function is of polynomial size w.r.t. AF. In particular, we identify that there is a direct relationship between the minimal models of a propositional formula and the preferred extensions of an argumentation framework by working on representing the defeated arguments. Then we show how to infer the preferred extensions of an argumentation framework by using UNSAT algorithms and disjunctive stable model solvers. The relevance of this result is that we define a direct relationship between one of the most satisfactory argumentation semantics and one of the most successful approach of non-monotonic reasoning i.e., logic programming with the stable model semantics.

The concept of the value-gradient is introduced and developed in the context of reinforcement learning, for deterministic episodic control problems that use a function approximator and have a continuous state space. It is shown that by learning the valuegradients, instead of just the values themselves, exploration or stochastic behaviour is no longer needed to find locally optimal trajectories. This is the main motivation for using value-gradients, and it is argued that learning the value-gradients is the actual objective of any value-function learning algorithm for control problems. It is also argued that learning value-gradients is significantly more efficient than learning just the values, and this argument is supported in experiments by efficiency gains of several orders of magnitude, in several problem domains. Once value-gradients are introduced into learning, several analyses become possible. For example, a surprising equivalence between a value-gradient learning algorithm and a policy-gradient learning algorithm is proven, and this provides a robust convergence proof for control problems using a value function with a general function approximator. Also, the issue of whether to include'residual gradient' terms into the weight update equations is addressed. Finally, an analysis is made of actor-critic architectures, which finds strong similarities to back-propagation through time, and gives simplifications and convergence proofs to certain actor-critic architectures, but while making those actor-critic architectures redundant. Unfortunately, by proving equivalence to policy-gradient learning, finding new divergence examples even in the absence of bootstrapping, and proving the redundancy of residual-gradients and actor-critic architectures in some circumstances, this paper does somewhat discredit the usefulness of using a value-function.

In an emergency situation, the actors need an assistance allowing them to react swiftly and efficiently. In this prospect, we present in this paper a decision support system that aims to prepare actors in a crisis situation thanks to a decision-making support. The global architecture of this system is presented in the first part. Then we focus on a part of this system which is designed to represent the information of the current situation. This part is composed of a multiagent system that is made of factual agents. Each agent carries a semantic feature and aims to represent a partial part of a situation. The agents develop thanks to their interactions by comparing their semantic features using proximity measures and according to specific ontologies.

Jerne's idiotypic network theory postulates that the immune response involves inter-antibody stimulation and suppression as well as matching to antigens. The theory has proved the most popular Artificial Immune System (ais) model for incorporation into behavior-based robotics but guidelines for implementing idiotypic selection are scarce. Furthermore, the direct effects of employing the technique have not been demonstrated in the form of a comparison with non-idiotypic systems. This paper aims to address these issues. A method for integrating an idiotypic ais network with a Reinforcement Learning based control system (rl) is described and the mechanisms underlying antibody stimulation and suppression are explained in detail. Some hypotheses that account for the network advantage are put forward and tested using three systems with increasing idiotypic complexity. The basic rl, a simplified hybrid ais-rl that implements idiotypic selection independently of derived concentration levels and a full hybrid ais-rl scheme are examined. The test bed takes the form of a simulated Pioneer robot that is required to navigate through maze worlds detecting and tracking door markers.

This paper examines the use of a hierarchical coevolutionary genetic algorithm under different partnering strategies. Cascading clusters of subpopulations are built from the bottom up, with higher-level subpopulations optimising larger parts of the problem. Hence higher-level subpopulations potentially search a larger search space with a lower resolution whilst lower-level subpopulations search a smaller search space with a higher resolution. The effects of different partner selection schemes amongst the subpopulations on solution quality are examined for two constrained optimisation problems. We examine a number of recombination partnering strategies in the construction of higher-level individuals and a number of related schemes for evaluating sub-solutions. It is shown that partnering strategies that exploit problemspecific knowledge are superior and can counter inappropriate (sub-) fitness measurements.

This paper presents a new type of genetic algorithm for the set covering problem. It differs from previous evolutionary approaches first because it is an indirect algorithm, i.e. the actual solutions are found by an external decoder function. The genetic algorithm itself provides this decoder with permutations of the solution variables and other parameters. Second, it will be shown that results can be further improved by adding another indirect optimisation layer. The decoder will not directly seek out low cost solutions but instead aims for good exploitable solutions. These are then post optimised by another hill-climbing algorithm. Although seemingly more complicated, we will show that this three-stage approach has advantages in terms of solution quality, speed and adaptability to new types of problems over more direct approaches. Extensive computational results are presented and compared to the latest evolutionary and other heuristic approaches to the same data instances. Introduction In recent years, genetic algorithms have become increasingly popular for solving complex optimisation problems such as those found in the areas of scheduling or timetabling. The general approach in the past was to directly optimise problems with a genetic algorithm often coupled with a post optimisation phase, i.e. both optimisation phases are directed towards lowering the cost of solutions. The new approach presented here is different in two respects. First, a separate decoding routine, with parameters provided by the genetic algorithm, solves the actual problem. Second, the aim of this decoder optimisation is not to achieve the lowest cost solutions in the first instance.

Contradiction is often seen as a defect of intelligent systems and a dangerous limitation on efficiency. In this paper we raise the question of whether, on the contrary, it could be considered a key tool in increasing intelligence in biological structures. A possible way of answering this question in a mathematical context is shown, formulating a proposition that suggests a link between intelligence and contradiction. A concrete approach is presented in the well-defined setting of cellular automata. Here we define the models of ``observer'', ``entity'', ``environment'', ``intelligence'' and ``contradiction''. These definitions, which roughly correspond to the common meaning of these words, allow us to deduce a simple but strong result about these concepts in an unbiased, mathematical manner. Evidence for a real-world counterpart to the demonstrated formal link between intelligence and contradiction is provided by three computational experiments.

The AAAI video archive is a central source of information about videotapes and films with information about AI that are stored digitally on other sites or physically in institutional archives. For each video, the archive includes a brief description of the contents and personae, one or more representative short clips for classroom or individual use, and the location of the archival copy (for example, at a university library).

Crosswords are very popular and represent a useful domain of investigation for modern artificial intelligence. In contrast to solving other celebrated games (such as chess), cracking crosswords requires a paradigm shift towards the ability to handle tasks for which humans require extensive semantic knowledge. This article introduces WebCrow, an automatic crossword solver in which the needed knowledge is mined from the web: clues are solved primarily by accessing the web through search engines and applying natural language processing techniques. In competitions at the European Conference on Artificial Intelligence (ECAI) in 2006 and other conferences this web-based approach enabled WebCrow to outperform its human challengers. Just as chess was once called “the Drosophila of artificial intelligence,” we believe that crossword systems can be useful Drosophila of web-based agents.

This article describes the architecture and AI technology behind an XML-based AI framework designed to streamline e-government form processing. The framework performs several crucial assessment and decision support functions, including workflow case assignment, automatic assessment, follow-up action generation, precedent case retrieval, and learning of current practices. To implement these services, several AI techniques were used, including rule-based processing, schema-based reasoning, AI clustering, case-based reasoning, data mining, and machine learning. The primary objective of using AI for e-government form processing is of course to provide faster and higher quality service as well as ensure that all forms are processed fairly and accurately. With AI, all relevant laws and regulations as well as current practices are guaranteed to be considered and followed. An AI framework has been used to implement an AI module for one of the busiest immigration agencies in the world.