Xianshun Chen

A Multi-Facet Survey on Memetic Computation

Memetic computation is a paradigm that uses the notion of meme(s) as units of information encoded in computational representations for the purpose of problem-solving. It covers a plethora of potentially rich meme-inspired computing methodologies, frameworks and operational algorithms including simple hybrids, adaptive hybrids and memetic automaton. In this paper, a comprehensive multi-facet survey of recent research in memetic computation is presented.

Hybrid ant colony algorithms for path planning in sparse graphs

The general problem of path planning can be modeled as a traveling salesman problem which assumes that a graph is fully connected. Such a scenario of full connectivity is however not always realistic. One such motivating example for us is the application of path planning for unmanned reconnaissance aerial vehicles (URAVs). URAVs are widely deployed for photography or imagery gathering missions of sites of interest. These sites can be targets in a combat zone to be investigated or sites inaccessible by ground transportation, such as those hit by forest fires, earthquake or other forms of natural disasters. The navigation environment is one where the overall configuration of the problem is a sparse graph. Unlike graphs that are fully connected, sparse graphs are not always Hamiltonian. In this paper, we describe hybrid ant colony algorithms (HACAs) proposed for path planning in sparse graphs since existing ant colony solvers designed for solving TSP do not apply to the present context directly. HACAs represent ant inspired algorithms incorporated with a local search procedure and some heuristic techniques for uncovering feasible route(s) or path(s) in a sparse graph within tractable time. Empirical results conducted on a set of generated sparse graphs demonstrate the excellent convergence property and robustness of HACAs in uncovering low risk and Hamiltonian visitation paths. Further, the obtained results also indicate that HACAs converge to secondary closed paths in situations where a Hamiltonian cycle does not exist theoretically or is not attainable within the bounded computational time window.

A Conceptual Modeling of Meme Complexes in Stochastic Search

In science, gene provides the instruction for making proteins, while meme is the sociocultural equivalent of a gene containing instructions for carrying out behavior. Taking inspiration from nature, we model the memeplex in search as instructions that specify the coadapted meme complexes of individuals in their lifetime. In particular, this paper presents a study on the conceptual modeling of meme complexes or memeplexes for more effective problem solving in the context of modern stochastic optimization. The memeplex representation, credit assignment criteria for meme coadaptation, and the role of emergent memeplexes in the lifetime learning process of a memetic algorithm in search are presented. A coadapted memetic algorithm that takes the proposed conceptual modeling of memeplexes into actions to solve capacitated vehicle routing problems (CVRPs) of diverse characteristics is then designed. Results showed that adaptive memeplexes provide a means of creating highly robust, self-configuring, and scalable algorithms, thus generating improved or competitive results when benchmarking against several existing adaptive or human-designed state-of-the-art memetic algorithms and metaheuristics, on a plethora of CVRP sets considered.

A self-adaptive memeplexes robust search scheme for solving stochastic demands vehicle routing problem

In this article, we proposed a self-adaptive memeplex robust search (SAMRS) for finding robust and reliable solutions that are less sensitive to stochastic behaviours of customer demands and have low probability of route failures, respectively, in vehicle routing problem with stochastic demands (VRPSD). In particular, the contribution of this article is three-fold. First, the proposed SAMRS employs the robust solution search scheme (RS 3) as an approximation of the computationally intensive Monte Carlo simulation, thus reducing the computation cost of fitness evaluation in VRPSD, while directing the search towards robust and reliable solutions. Furthermore, a self-adaptive individual learning based on the conceptual modelling of memeplex is introduced in the SAMRS. Finally, SAMRS incorporates a gene-meme co-evolution model with genetic and memetic representation to effectively manage the search for solutions in VRPSD. Extensive experimental results are then presented for benchmark problems to demonstrate that the proposed SAMRS serves as an efficable means of generating high-quality robust and reliable solutions in VRPSD.

Agent-Based Modeling and Simulation for Supply Chain Risk Management - A Survey of the State-of-the-Art

Due to the complex nature and numerous interacting factors that contributes to the increased vulnerability of supply chains, traditional methods have been found to be inadequate for Supply Chain Risk Management (SCRM). Agent-Based Modeling and Simulation (ABMS), an agent-oriented approach to model and simulate complex adaptive systems, represents a recent development in supply chain planning that has been regarded highly appropriate for studying risk management [1-4]. The objective of this paper is to provide a multi-perspective survey of the state-of-the-art agent based modeling and simulation approaches for SCRM.

Autonomous multi-agents in flexible flock formation

In this paper, we present a flocking model where agents are equipped with navigational and obstacle avoidance capabilities that conform to user defined paths and formation shape requirements. In particular, we adopt an agent-based paradigm to achieve flexible formation handling at both the individual and flock level. The proposed model is studied under three different scenarios where flexible flock formations are produced automatically via algorithmic means to: 1) navigate around dynamically emerging obstacles, 2) navigate through narrow space and 3) navigate along path with sharp curvatures, hence minimizing the manual effort of human animators. Simulation results showed that the proposed model leads to highly realistic, flexible and real-time reactive flock formations.

FAME, soft flock formation control for collective behavior studies and rapid games development

We present FAME, a comprehensive C# software library package providing soft formation control for large flocks of agents. While many existing available libraries provide means to create flocks of agent equipped with simple steering behavior, none so far, to the best of our knowledge, provides an easy and hassle free approach to control the formation of the flock. Here, besides the basic flocking mechanisms, FAME provides an extensive range of advanced features that gives enhanced soft formation control over multiple flocks. These soft formation features include defining flocks in any user-defined formation, automated self-organizing agent within formation, manipulating formation shape at real-time and bending the formation shape naturally along the curvature of the path. FAME thus not only supports the research studies of collective intelligence and behaviors, it is useful for rapid development of digital games. Particularly, the development cost and time pertaining to the creation of multi-agent group formation can be significantly reduced.

Conceptual modeling of evolvable local searches in memetic algorithms using linear genetic programming: a case study on capacitated vehicle routing problem

This paper presents a study on the conceptual modeling of memetic algorithm with evolvable local search in the form of linear programs, self-assembled by linear genetic programming based evolution. In particular, the linear program structure for local search and the associated local search self-assembling process in the lifetime learning process of memetic algorithm are proposed. Results showed that the memetic algorithm with evolvable local search provides a means of creating highly robust, self-configuring and scalable algorithms, thus generating improved or competitive results when benchmarking against several existing adaptive or human-designed state-of-the-art memetic algorithms and meta-heuristics, on a plethora of capacitated vehicle routing problem sets considered.

Interactive GA flock brush for non-photorealistic rendering

Art styles are modes of expressing creative artistic ideas. Non-photorealistic rendering is a process of projecting artistic expressions in a digital representation. In this paper, we consider the use of evolutionary computation techniques to explore the variability of artistic styles through an evolutionary process. Our system, a union of biological swarms in the form of flocks and interactive genetic algorithm (IGA), generates artistic styles to produce stylized digital photographs. By varying a finite set of parameters, we transform photo-realistic scenes to artistic imagery. Our most distinct styles bear close resemblance to familiar traditional art styles like Impressionism and Pointillism.

Autonomous flock brush for non-photorealistic rendering

Non-photorealistic rendering systems strive to create compelling stylized effects from realistic images. We present an interactive process using flocks of autonomous agents to model a painters brush. As flocks of agents glide across the canvas like bristles on a paint brush, a stylized picture can be produced by carefully directing the path of movement. The agents leave behind a trail of color resulting in painterly or pencil sketch looking images.