According to Dynamic Logic Programming (DLP), knowledge may be given by a set of theories (encoded as logic programs) representing different states of knowledge. These may represent time (in updates), specificity (in taxonomies), strength of updating instance (in the legislative domain), hierarchical position of knowledge source (in organizations), etc. The mutual relationships extant among states are used to determine the semantics of the combined theory composed of all the individual theories. Although suitable to encode a single dimension (e.g. time, hierarchies...), DLP cannot deal with more than one simultaneously because it is defined only for a linear sequence of states. To overcome this limitation, we introduce the notion of Multi-dimensional Dynamic Logic Programming (MDLP), which generalizes DLP to collections of states organized in arbitrary acyclic digraphs representing precedence. In this setting, MDLP assigns semantics to sets and subsets of such logic ...

This paper explores the applicability of the new paradigm of Multi-dimensional Dynamic Logic Programming to represent an agent’s view of the combination of societal knowledge dynamics. The representation of a dynamic society of agents is the core of MIN ERVA [11], an agent architecture and system designed with the intention of providing a common agent framework based on the unique strengths of Logic Programming, hat allows the combination of several non-monotonic knowledge representation and reasoning mechanisms developed in recent years.

We focus on learning representations of dynamical systems that can be characterized by logic-based formalisms for reasoning about actions and change, where system's behaviors are naturally viewed as appropriate logical consequences of the domain's description. To this end, logic-based induction methods are adapted to identify the input/output behavior of a dynamical system corresponding to an environment. The study of dynamic domains is started with domains modelable with classical action theories and is progressively enhanced to manage more complex behaviors.

Inductive Logic Programming (ILP) is concerned with the task of generalising sets of positive and negative examples with respect to background knowledge expressed as logic programs. Negation as Failure (NAF) is a key feature of logic programming which provides a means for nonmonotonic commonsense reasoning under incomplete information. But, so far, most ILP research has been aimed at Horn programs which exclude NAF, and has failed to exploit the full potential of normal programs that allow NAF. By contrast, Abductive Logic Programming (ALP), a related task concerned with explaining observations with respect to a prior theory, has been well studied and applied in the context of normal logic programs. This paper shows how ALP can be used to provide a semantics and proof procedure for nonmonotonic ILP that utilises practical methods of language and search bias to reduce the search space. This is done by lifting an existing method called Hybrid Abductive Inductive Learning (HAIL) from Horn clauses to normal logic programs. To demonstrate its potential benefits, the resulting system, called XHAIL, is applied to a process modelling case study involving a nonmonotonic temporal Event Calculus (EC). 1

Abstract not found

This paper explores the applicability of the new paradigm of Multi-dimensional Dynamic Logic Programming to represent an agent's view of the combination of societal knowledge dynamics. The representation of a dynamic society of agents is the core of A4ZJVTI/A [11], an agent architecture and system designed with the intention of providing a common agent framework based on the unique strengths of ogic Programming, hat allows the combination of several non-monotonic knowledge representation and reasoning mechanisms developed in recent years.

We present a logical framework and the declarative semantics of a multi-agent system in which each agent can communicate with and update other agents, can update its knowledge state, of its own initiative or when it receives new incoming information, and it is able to prefer beliefs and reactions, when several alternatives are possible. The knowledge state of an agent is represented by an updatable prioritized logic program, in which priorities among rules can be expressed to allow the agent to prefer, and where the preference relation itself can be updated as well. An example is developed to illustrate how our approach works, including how preferring can enhance reactivity in agents. Finally, we discuss some Web applications where our approach can have a significant potential to contribute, and refer to the work of others.

Introduction In a previous paper [5] we presented a combination of the dynamic logic programming paradigm proposed by J. J. Alferes et al. [1, 10] and a version of KS-agents proposed by Kowalski and Sadri [7]. In the resulting framework, rational, reactive agents can dynamically change their own knowledge bases as well as their own goals. In particular, at every iteration of an observe-think-act cycle, the agent can make observations, learn new facts and new rules from the environment, and then it can update its knowledge accordingly. The agent can also receive a piece of information that contrasts with its knowledge. To solve eventual cases of contradiction within the theory of an agent, techniques of contradiction removal and preferences among several sources can be adopted [8]. The actions of an agent are modeled by means of updates, inspired by the approach in [3]. A semantic characterization of updates is given in [1] as a generalization of the stable model semantics of n

We present a method to learn simultaneously definitions for a concept and its negation. This problem is relevant when we have to deal with a complex domain where it is difficult to acquire a complete theory and where we have to reason from incomplete knowledge. We use default logic to represent such incomplete theories. This paper specifies the problem of learning a default theory from a set of examples and a background knowledge. We propose an operational method to inductively construct such a theory. Our learning process relies on a generalization mechanism defined in the field of Inductive Logic Programming. We first consider the case where the initial knowledge is sure because it contains only ground facts. Then, we extend the framework to the case where the initial knowledge is a default theory.

Abstract. Inductive logic programming is a form of machine learning from examples which employs the representation formalism of clausal logic. One of the earliest inductive logic programming systems was Ehud Shapiro’s Model Inference System [90], which could synthesise simple recursive programs like append/3. Many of the techniques devised by Shapiro, such as top-down search of program clauses by refinement operators, the use of intensional background knowledge, and the capability of inducing recursive clauses, are still in use today. On the other hand, significant advances have been made regarding dealing with noisy data, efficient heuristic and stochastic search methods, the use of logical representations going beyond definite clauses, and restricting the search space by means of declarative bias. The latter is a general term denoting any form of restrictions on the syntactic form of possible hypotheses. These include the use of types, input/output mode declarations, and clause schemata. Recently, some researchers have started using alternatives to Prolog featuring strong typing and real functions, which alleviate the need for some of the above ad-hoc mechanisms. Others have gone beyond Prolog by investigating learning tasks in which the hypotheses are not definite clause programs, but for instance sets of indefinite clauses or denials, constraint logic programs, or clauses representing association rules. The chapter gives an accessible introduction to the above topics. In addition, it outlines the main current research directions which have been strongly influenced by recent developments in data mining and challenging real-life applications. 1