Abstract. Supervised learning in attribute-based spaces is one of the most popular machine learning problems studied and, consequently, has attracted considerable attention of the genetic algorithm community. The fullmemory approach developed here uses the same nigh-level descriptive language that is used in rule-based systems. This allows for an easy utilization of inference rules of the well-known inductive learning methodology, which replace the traditional domain-independent operators and make the search task-specific. Moreover, a closer relationship between the underlying task and the processing mechanisms provides a setting for an application of more powerful task-specific heuristics. Initial results obtained with a prototype implementation for the simplest case of single concepts indicate that genetic algorithms can be effectively used to process nigh-level concepts and incorporate task-specific knowledge. The method of abstracting the genetic algorithm to the problem level, described here for the supervised inductive learning, can be also extended to other domains and tasks, since it provides a framework for combining recently popular genetic algorithm methods with traditional problem-solving methodologies. Moreover, in this particular case, it provides a very powerful tool enabling study of the widely accepted but not so well understood inductive learning methodology.

Evolution is a fundamental form of adaptation in a dynamic and complex environment. Genetic algorithms are an effective tool in the empirical study of evolution. This paper follows Axelrod's work [2] in using the genetic algorithm to evolve strategies for playing the game of Iterated Prisoner's Dilemma, using co-evolution, where each member of the population (each strategy) is evaluated by how it performs against the other members of the current population. This creates a dynamic environment in which the algorithm is optimising to a moving target instead of the usual evaluation against some fixed set of strategies. The hope is that this will stimulate an "arms race" of innovation [3]. We conduct two sets of experiments. The first set investigates what conditions evolve the best strategies. The second set studies the robustness of the strategies thus evolved, that is, are the strategies useful only in the round robin of its population or are they effective against a wide variety of oppo...

Abstract not found

SAMUEL is an experimental learning system that uses genetic algorithms and other learning methods to evolve reactive decision rules from simulations of multi-agent environments. The basic approach is to explore a range of behavior within a simulation model, using feedback to adapt its decision strategies over time. One of the main themes in this research is that the learning system should be able to take advantage of existing knowledge where available. This has led to the adoption of rule representations that ease the expression existing knowledge. A second theme is that adaptation can be driven by competition among knowledge structures. Competition is applied at two levels in SAMUEL. Within a strategy composed of decision rules, rules compete with one another to influence the behavior of the system. At a higher level of granularity, entire strategies compete with one another, driven by a genetic algorithm. This article focuses on recent elaborations of the agent model of SAMUEL that a...

In this paper, we use a genetic algorithm to evolve a set of classification rules with real-valued attributes. We show how real-valued attribute ranges can be encoded with real-valued genes and present a new uniform method for representing don't cares in the rules. We view supervised classification as an optimization problem, and evolve rule sets that maximize the number of correct classifications of input instances. We use a variant of the Pitt approach to genetic-based machine learning system with a novel conflict resolution mechanism between competing rules within the same rule set. Experimental results demonstrate the effectiveness of our proposed approach on a benchmark wine classifier system. I. Introduction Genetic algorithms (GAs) have proved to be robust, domain independent mechanisms for numeric and symbolic optimization[7]. Our previous work has demonstrated effective genetic-based rule learning in discrete domains [13]. In the real world, however, most classification prob...

This paper describes a natural language system which improves its own performance through learning. The system processes short English narratives and is able to acquire, from a single narrative, a new schema for a stereotypical set of actions. During the understanding process, the system attempts to construct explanations for characters ' actions in terms of the goals their actions were meant to achieve. When the system observes that a character has achieved an interesting goal in a novel way, it generalizes the set of actions they used to achieve this goal into a new schema. The generalization process is a knowledge-based analysis of the causal structure of the narrative which removes unnecessary details while maintaining the validity of the causal explanation. The resulting generalized set of actions is then stored as a new schema and used by the system to correctly process narratives which were previously beyond its capabilities. I

Cooperative co--evolutionary systems can facilitate the development of teams of heterogeneous agents. We believe that k different behavioral strategies for controlling the actions of a group of k agents can combine to form a cooperation strategy which efficiently achieves global goals. We examine the on--line adaption of behavioral strategies utilizing genetic programming. Specifically, we deal with the credit assignment problem of how to fairly split the fitness of a team to all of its participants. We present several crossover mechanisms in a genetic programming system to facilitate the evolution of more than one member in the team during each crossover operation. Our goal is to reduce the time needed to evolve a good team. 1 Introduction We have utilized genetic programming (GP) [ Koza, 1992 ] to evolve behavioral strategies which enabled a team of loosely--coupled agents to cooperatively achieve a common goal [ Haynes and Sen, 1996, Haynes et al., 1995 ] . Since they each shared ...

We introduce a cooperative co--evolutionary system to facilitate the development of teams of heterogeneous agents. We believe that k different behavioral strategies for controlling the actions of a group of k agents can combine to form a cooperation strategy which efficiently achieves global goals. We both examine the on-line adaption of behavioral strategies utilizing genetic programming and demonstrate the successful co-evolution of cooperative individuals. We present a new crossover mechanism for genetic programming systems in order to facilitate the evolution of more than one member in the team during each crossover operation. Our goal is to reduce the time needed to evolve an effective team.

One approach to the design of learning systems is to extract heuristics from existing adaptive systems. Genetic algorithms are heuristic learning models based on principles drawn from natural evolution and selective breeding. Some features that distinguish genetic algo- rithms from other search methods are: A population of structures that can be interpreted as candidate solutions to the given problem; The competitive selection of structures for reproduction, based on each structure's fitness as a solution to the given problem; Idealized genetic operators that alter the selected structures in order to create new structures for fur- ther testing.

Collective action has been examined to expedite search in optimization problems [ Dorigo et al., 1996 ] . Collective memory has been applied to learning in multiagent systems [ Garland and Alterman, 1996 ] . We integrate the simplicity of collective action with the pattern detection of collective memory to significantly improve both the gathering and processing of knowledge. We investigate the augmentation of distributed search in genetic programming based systems with collective memory. Four models of collective memory search are defined based on the interaction of the search agents and the process agents which manipulate the collective memory. We present implementations of two of the collective memory search models and further show how collective memory search facilitates "scaling up" a problem domain. An Active-Passive model, which gathers results from the independent searchers, is examined and found to provide a springboard from which search agents can extend their exploration. A P...