In this paper, a fitness landscape analysis for several instances of the quadratic assignment problem (QAP) is performed and the results are used to classify problem instances according to their hardness for local search heuristics and meta-heuristics based on local search. The local properties of the tness landscape are studied by performing an autocorrelation analysis, while the global structure is investigated by employing a fitness distance correlation analysis. It is shown that epistasis, as expressed by the dominance of the flow and distance matrices of a QAP instance, the landscape ruggedness in terms of the correlation length of a landscape, and the correlation between fitness and distance of local optima in the landscape together are useful for predicting the performance of memetic algorithms - evolutionary algorithms incorporating local search - to a certain extent. Thus, based on these properties a favorable choice of recombination and/or mutation operators can be found.

This chapter addresses embodied social interaction inlif6 like agents. Embodiment is discussedf rom both arti cial intelligence and psychology viewpoints. Di#erent degreesof embodiment in biological, virtual and robotic agents are discussed, given the example of a bottomup, behavior-oriented, dynamic control of virtual robots. A `dancing with strangers' experiment shows how the same principles can be applied to physical robot-human interaction. We then discuss the issue of sociality which di#ers in di#erent academic communities with respect to which roles are attributed to genes, memes, and the individual embodied agent.

In this paper we introduce a promising hybridization scheme for a Memetic Algorithm (MA). Our MA is composed of two optimization processes, a Genetic Algorithm and a Monte Carlo method (MC). In contrast with other GA-Monte Carlo hybridized memetic algorithms, in our work the MC stage serves two purposes: -- when the population is diverse it acts like a local search procedure and -- when the population converges its goal is to diversify the search. To achieve this, the MC is self-adaptive based on observations from the underlying GA behavior; the GA controls the long-term optimization process. We present preliminary, yet statistically significant, results on the application of this approach to the TSP problem.We also comment it successful application to a molecular conformational problem: Protein Folding.

This article puts research on socially intelligent agents (SIA) in the broader context of how humans (and other primates) perceive and interact with the social world. Phylogenetic (evolutionary) and ontogenetic (developmental) issues are discussed with respect to the social origin of primate and human intelligence and human culture. Implications for designing artifacts and for the evolvability of human societies are outlined. A theory of empathy is presented that is based on current research on the primate social brain. Research projects that investigate some of these issues are reviewed. I argue that Socially Intelligent Agents (SIA) research, although strongly linked to software and robotic engineering, goes beyond a software engineering paradigm: it can potentially serve as a paradigm for a science of social minds. A systematic and experimental investigation of human social minds and the way humans perceive the social world can result in truly social artifacts,...

We present a conceptual framework for the study of mate choice in monogamous mating systems with non-negligible courtship time. Within this framework, we develop a mate choice model for the common case where individuals have a changing social network of potential partners. We evaluate

Microarrays have become a key technology in experimental molecular biology. They allow a monitoring of gene expression for more than ten thousand genes in parallel producing huge amounts of data. In the exploration of...

We present a system for performing belief revision in a multi-agent environment. The system is called GBR (Genetic Belief Revisor) and it is based on a genetic algorithm. In this setting, different individuals are exposed to different experiences. This may happen because the world surrounding an agent changes over time or because we allow agents exploring different parts of the world. The algorithm permits the exchange of chromosomes from different agents and combines two different evolution strategies, one based on Darwin's and the other on Lamarck's evolutionary theory. The algorithm therefore includes also a Lamarckian operator that changes the memes of an agent in order to improve their fitness. The operator is implemented by means of a belief revision procedure that, by tracing logical derivations, identifies the memes leading to contradiction. Moreover, the algorithm comprises a special crossover mechanism for memes in which a meme can be acquired from another agent only if the other agent has "accessed" the meme, i.e...

A vast number of very successful applications of Memetic algorithms (MAs) have been reported in the literature in the last years for a wide range of problem domains. The majority of the papers dealing with MAs are the result of the combination of highly specialised preexisting local searchers and usually purpose-speci c genetic operators. Moreover, those algorithms require a considerable eort devoted to the tuning of the local search and evolutionary parts of the algorithm. We have demonstrated in our previous work (see references below), that given a range of possible local search strategies available to a Memetic Algorithm, the optimal choice of which one must be used is not only problem and instance dependent but also tightly related to the state of the search process itself. We also showed that it is indeed possible to produce Memetic Algorithms that adapt on-the-y to those situations for a variety of problem domains. In this paper we continue our studies of the design of robust Memetic Algorithms by introducing the concept of \self-generating" Memetic Algorithms. As mentioned above the success of a Memetic Algorithm depends on the pre-existence of powerful local searchers. Here we allow the Memetic Algorithm to create its local searchers and to co-evolve the behaviours it needs to successfully solve a problem. 1

We present a simulation of a society of agents where some of them have \moral sentiments " towards the agents that belong to the same social group, using the Iterated Prisoner 's Dilemma as a metaphor for the social interactions. Besides the well-understood phenomenon of short-sighted, self-interested agents performing well in the short-term but ruining their chances of such performance in the long run in a world of reciprocators, the results suggest that, where some agents are more generous than that, these agents have a positive impact on the social group to which they belong, without compromising too much their individual performance (i.e., the group performance improves). The inspiration for this project comes from a discussion on Moral Sentiments by M.Ridley. We describe various simulations where conditions and parameters over determined dimensions were arranged to account for dierent types and compositions of societies. Further, we indicate several lessons that arise from the analysis of the results and comparison of the dierent experiments. We also relate this work to our previous anthropological approach to the adaptation of migrant agents, and argue that allowing agents to possess suitably-chosen emotions can have a decisive impact on Multi-Agent Systems. This implies that some common notions of agent autonomy (and related concepts) should be reexamined.

The paper discusses an evolutionary knowledge approach to intelligent problem solving. A rule-based production system is used to model the problem and the means by which the problem space should be searched.