Entropy-based descriptors are widely used for two-dimensional texture irregularity analysis. In counting-based formulations such as two-dimensional Dispersion Entropy ($\text{DispEn}\_{2\mathrm{D}}$), potential pattern counts grow rapidly with embedding dimension while available templates remain limited by image size, which can bias probability estimation and compress responses. Mosaic Entropy ($\text{MosEn}\_{2\mathrm{D}}$) is proposed, combining binary mosaic representation with conditional probability counting to stabilize statistics under limited template support. Experiments on$\text{WGN}\_{2\mathrm{D}}$,$\text{MIX}\_{2\mathrm{D}}(p)$, Kylberg textures, and Kylberg ISO noise show that$\text{MosEn}\_{2\mathrm{D}}$improves image-size stability, reduces saturation as irregularity increases, and preserves distinguishable responses under realistic sensor noise, with lower computational overhead. It also maintains texture discrimination comparable to$\text{DispEn}\_{2\mathrm{D}}$without requiring an additional class parameter.

This position paper contends that as artificial intelligence (AI) and robotics advance at a rapid pace—transforming industries, healthcare, transportation, and more—the development of structured knowledge and formal ontologies struggles to keep up. This imbalance poses substantial risks, as increasingly powerful and autonomous systems can make decisions that are both opaque and difficult to verify. While AI and robotics research benefit from significant funding and widespread attention, ontology and structured knowledge efforts often remain under-resourced, under-integrated, and often sidelined—leaving a growing semantic gap in system design and governance. To address this gap, we argue that formal ontology research must be reinvigorated to keep pace with the accelerating demands of AI and robotics—not merely as a support function but as a core contributor to trustworthy and beneficial AI/robotics system design. This requires renewed investment in the academic foundations of ontology engineering, reintegration of semantic modeling into AI/robotics development workflows, and the lowering of practical barriers that have hindered broader adoption. Specifically, we advocate for ontology-augmented verification methods that incorporate semantic constraints into behavioral validation; for the institutionalization of semantic auditing practices that allow ontologies to serve as transparent, inspectable system referents; and for the adoption of FAIR publishing standards that treat ontologies as first-class research outputs. These directions, we argue, are essential to bridge the gap between AI/robotics and ontology research and to ensure that autonomous AI/robotics systems remain not only performant but also transparent, accountable, and ultimately beneficial to humanity.

Generative AI has achieved remarkable progress in automating creative processes, from generating images and composing music to crafting textual narratives. However, a significant gap remains between generative AI and human creativity. While current systems excel in pattern replication, they lack the dynamic exploration, contextual awareness, and iterative refinement that are hallmarks of human creativity. This paper advocates for a paradigm shift in generative AI, emphasizing the integration of context-aware mechanisms and dynamic evaluation processes to better emulate the adaptive and exploratory nature of human creative workflows. To better articulate this position, we outline a meta-framework that draws on existing methods in optimization and knowledge representation—such as evolutionary computation, ontologies, and machine learning—to conceptualize how aesthetic hunger may be cultivated in generative systems, allowing them to actively search, refine, and evolve creative expressions, mirroring human artistic exploration. Using music composition as a guiding example, we illustrate how generative systems can exhibit adaptability and exploratory behaviors, akin to human creators, in navigating uncharted aesthetic spaces and refining their artistic visions. By further fostering interdisciplinary collaborations, particularly with artists and designers, this approach not only expands AI capabilities but also enriches our understanding of human creativity, reaffirming its central role in shaping technological and cultural landscapes.

This paper enhances option pricing accuracy by incorporating financial expertise into a neural network (NN) design and optimizing data sample quality through cleaning and synthesis. Instead of directly estimating option values (OVs) with NNs, we leverage the concept of control variate by decomposing OVs as time values (TVs) estimated by NNs, plus the analytically solvable intrinsic values (IVs). TV surface can be decomposed into two scenarios with very different properties, and we design two NNs according to our derived no-arbitrage constraints for these two scenarios. To alleviate learning inaccuracy due to the kink of the TV surface along the scenario boundary, we synthesize training samples based on our derived constraints to smoothly extend the surface for each scenario. On the other hand, irrational option quotes commonly found in illiquid markets incur uneven surfaces, significantly deteriorating NN predictability. We develop a learnable data-cleaning method to remove potentially irrational quotes spotted by no-arbitrage constraints properly. Besides, unnecessary data syntheses proposed in previous literature can also be removed by incorporating corresponding constraints into our NN to enhance training efficiency. Comprehensive experiments on liquid S&P 500 and illiquid TAIEX option markets examine the superiority of our approach.

Based on a previously proposed meta-framework, called ants on multiple graphs (AntsOMG), this paper further investigates into certain more complicated creative behavior, in which creators interact with their own creations in order to complete works and elevate the creation forms to a higher level, via computational mechanisms. Aiming to achieve the goal of this study, the target music genre, organum motets, is adopted because its composition process requires iterative interactions with composed music and quite resembles the creative behavior under investigation. AntsOMG is employed as a fundamental component for developing models and conducting creations according to the developed models. With the assistance of AntsOMG, the proposed approach firstly develops style models for section scheme planning and accordingly plans section schemes for the organum motets to be composed. Secondly, following the section scheme, cantus firmus is composed and also used to generate the graph models to be developed into the style models required for composing the corresponding contrapuntal parts for the polyphonic sections of the organum motets. For finalizing each contrapuntal section, a genetic algorithm is utilized to introduce variety and diversity. The outcomes demonstrate that contrapuntal composition and autonomous style development of organum motets can be achieved by the proposed approach. The contribution of this paper is twofold. First, the presented implementation is immediately applicable to compose unlimited amount of organum motets, which may not be possible for human composers. Second, the success of the proposal may shed light on gaining further understandings of complicated creative behavior.

In this paper, genetic algorithms are proposed to shrink counterexamples found by QuickChick, a property-based testing framework for Coq. In order to make the outcome of property-based testing humanly understandable and inspectable, genetic algorithms are brought into the realm of rigorous software development as shrinkers capable of handling a broad range of data structures. In the present study, two showcases, merge sort and insertion of red-black trees, are investigated for illustrative purposes. Due to the lack of relevant results existing in the literature, two baseline methods, random sample and random walk are included in the experiments for comparison with the proposed genetic algorithm. The obtained results indicate that the proposal is effective since the program mistake can be identified with ease by examining the shrunk counterexamples and also that the adopted genetic algorithm statistically significantly outperforms random sample and random walk in both counterexample sizes and running time.

This paper presents a study on fusing creative operations into evolutionary computation for music composition, as an attempt to provide a collaborative creation environment between human and machine and to investigate into the mind process during music creation. A framework based on the paradigm of evolutionary computation is proposed to incorporate key mechanisms observed in the creation process and implemented to work on evolving abstract musical ideas and thoughts. The developed software framework is released as open source. The collaborative creation process, involving the framework, the composer, and the performer, is presented and discussed, and the results of this study also include a composed music work for unaccompanied cello, played by a renowned cellist.

The link between logic and computation has been established by the BHK interpretation and the Curry-Howard isomorphism, based on which proof assistants capable of verifying formal proofs by transforming proofs into programs and by computationally evaluating the programs have been developed for the past few decades. Because evolutionary algorithms are search methods with remarkable feasibility and can be used to automatically generate programs, in our previous proposal, evolutionary algorithms and proof assistants were integrated to create a framework able to automatically prove simple theorems. In the present work, we aim to enhance the search ability of the proof generator such that proofs of slightly advanced, complicated theorems can be generated via evolutionary search with human involvement. This article describes in detail the algorithmic design of the proposed proof generator, how and why humans are involved in the process of proof development, and the test runs, in which proofs as Coq formalization of three theorems, the divisibility rule for 3, the sum of an arithmetic series, and the inequality of arithmetic and geometric means, were successfully generated. The developed source code with the obtained experimental results, including the human created rules and the software generated proofs, are released as open source.

Mathematical theorems are human knowledge able to be accumulated in the form of symbolic representation, and proving theorems has been considered intelligent behavior. Based on the BHK interpretation and the Curry-Howard isomorphism, proof assistants, software capable of interacting with human for constructing formal proofs, have been developed in the past several decades. Since proofs can be considered and expressed as programs, proof assistants simplify and verify a proof by computationally evaluating the program corresponding to the proof. Thanks to the transformation from logic to computation, it is now possible to generate or search for formal proofs directly in the realm of computation. Evolutionary algorithms, known to be flexible and versatile, have been successfully applied to handle a variety of scientific and engineering problems in numerous disciplines for also several decades. Examining the feasibility of establishing the link between evolutionary algorithms, as the program generator, and proof assistants, as the proof verifier, in order to automatically find formal proofs to a given logic sentence is the primary goal of this study. In the article, we describe in detail our first, ad-hoc attempt to fully automatically prove theorems as well as the preliminary results. Ten simple theorems from various branches of mathematics were proven, and most of these theorems cannot be proven by using the tactic auto alone in Coq, the adopted proof assistant. The implication and potential influence of this study are discussed, and the developed source code with the obtained experimental results are released as open source.

MapReduce is a popular distributed programming framework for large-scale data processing. To prevent MapReduce jobs from being interrupted by node failures that occur frequently in a MapReduce cluster consisting of a set of commodity machines/nodes, the most well-known MapReduce implementation, i.e. Hadoop, adopts a task re-execution policy (TR policy). When a map/reduce task of a job crashes, the TR policy assigns another node to reperform the task. However, the impact of the TR policy on MapReduce jobs in terms of reliability, job turnaround time (JTT) and energy consumption are not clear, particularly when jobs have different features, e.g. different filtering percentages, different input-data sizes, and different numbers of reduce tasks. In this paper, we formally analyze the job completion reliability (JCR) of a job based on Poisson distributions, and then derive the expected JTT and job energy consumption (JEC) based on the universal generation function. Extensive analyses are further conducted to explore the impact of the TR policy on JCR, JTT and JEC of jobs with different features. The results show that employing the TR policy can dramatically improve JCR for a large MapReduce job. Moreover, if the JCR of a job is highly improved by the TR policy, the expected JTT and JEC will not be significantly prolonged and increased, respectively.

Scientists and artists have been dedicate themselves to combine science and art together for decades, especially in the realm of computer science and digital technology. Nowadays, there are commercial software packages providing artistic image effect filters and processors. These functionalities are usually designed for certain specific purposes, such as blur, sketch, and the like. In this study, we make our attempt to develop a mechanism which is able to generate many different kinds of artistic effects, many of which are probably difficult to describe. The proposed mechanism combines cellular automata and pixel painting to process a given image with the involvement of an evolutionary algorithm. Interesting and/or unexpected results are obtained via the use of the evolutionary algorithm on the objective functions presented in this paper.

In this paper we propose a novel co-allocation scheme, called a Reliable Multicast Brute-Force co-allocation scheme (ReMBF for short), which employs a reliable multicast (RM for short) technique with the Brute-Force (BF for short) scheme to accelerate data retrieval and delivery, and reliably transmit data to its users for data grids. Several types of data access patterns, including Zipf-like, geometric, and uniform distributions, are utilized to model user access behaviors and evaluate the performance of ReMBF. The simulation results demonstrate that ReMBF can efficiently deliver a bulk of data in a shorter time period compared with two state-of-the-art schemes.

Recently, MapReduce has been widely employed by many companies/organizations to tackle data-intensive problems over a large-scale MapReduce cluster. To solve machine/node failure which is inevitable in a MapReduce cluster, MapReduce employs several policies, such as input-data replication and intermediate-data replication policies. To speed up job execution, MapReduce allows reduce tasks to early fetch their required intermediate data. However, the impact of these policy combinations on the job completion reliability (JCR for short) and job energy consumption (JEC for short) of a MapReduce cluster was not clear, where JCR is the reliability with which a MapReduce job can be completed by the cluster, whereas JEC is the energy consumed by the cluster to complete the job. Therefore, in this study, we analyze the JCR and JEC of a MapReduce cluster on four policy combinations (POCs for short) derived from two typical intermediate-data replication policies and two typical reduce-task assignment policies. The four POCs are further compared in extensive scenarios, which not only consider jobs at different scales with various parameters, but also give a MapReduce cluster two extreme parallel execution capabilities and diverse bandwidths. The analytical results enable MapReduce managers to comprehend how these POCs impact the JCR and JEC of a cluster and then select an appropriate POC based on the characteristics of their own MapReduce jobs and clusters.

In recent years, many social network systems like, YouTube, Facebook, Twitter, etc. have been a part of our everyday life. Among these systems, YouTube which plays video programs of different interesting themes for users has been one of the most attractive ones. Basically, when the space of Memcached in YouTube is full, the Least Recently Used algorithm (LRU for short) is employed to evict a least recently watched video. However, the LRU, due to its own property, may cause more miss counts of Memcached for YouTube, consequently increasing the network bandwidth and energy consumptions. To solve these problems, in this paper we proposed two cache replacement algorithms, the Pareto Least Recently Used algorithm (PLRU for short) and Pareto Least Frequently Used algorithm (PLFU for short), in which videos are classified into different popularity categories, and those videos in the top 10% and 20% of the top two popular categories of YouTube, based on Pareto principle, are then chosen to serve users' requests without removing them from Memcached. The simulation results show that the PLFU algorithm can significantly reduce miss counts of Memcached compared with those when the LRU and Least Frequently Used algorithm (LFU for short) are employed, thus achieving great performance improvements for the top two popular categories of videos. Also, when a large space of Memcached is available, the miss counts of the PLRU are higher than those of the LRU. But when comparing them in a small available Memcached space, the miss counts of the LRU on the contrary is higher than those of the PLRU.

MapReduce has been a worldwide accepted framework for solving data-intensive applications. To prevent MapReduce jobs from being interrupted by node failures which occur frequently in a large-scale MapReduce cluster, current MapReduce implementations, e.g., Hadoop, employ a task re-execution policy (TR policy for short) for MapReduce jobs, i.e., when a map/reduce task of a job fails due to node failure, this policy reperforms the task on another node. However, the impact of the TR policy on job completion reliability and job completion time have not been studied from a theoretical viewpoint, especially when the job is given different characteristics, e.g., different input data sizes, different numbers of reduce tasks, and different intermediate data sizes. In this study, we derive the job completion reliability (JCR for short) of a MapReduce job based on Poisson distributions and analyze the expected job completion time (JCT for short) based on the universal generation function. We use nine settings of task re-execution factor (TR factor for short) to explore the impact of the TR policy on the JCR and JCT of jobs. The results show that the TR policy can effectively improve JCR without significantly prolonging JCT. But there is no single TR factor with which all jobs can achieve a high JCR.

Evacuation simulation is a critical and important research issue for people to design safer building layouts or plan more effective evacuation routes. Many studies adopted methodologies in evolutionary computation into the evacuation simulation systems for finding better solutions. To simulate human behavior or crowd motion is one key factor to the practicality of the system. Particle swarm optimization algorithm (PSO), which is originated from the inspiration of bird flocking, is commonly applied to model human behavior. Based on the PSO-based human behavior simulation, many studies have got good results on evacuation simulation. However, the configurations of describing the experiment environment in the literature are complicated and specialized for certain specific scenarios. Observing the fact, we propose a new PSO-based simulation framework in order to provide a simple and general way to configure various simulation scenarios. This work adopts our previously proposed PSO-based crowd movement controlling mechanism and introduces new mechanisms to make the simulation fitting into evacuation circumstance more real. In the proposed framework, all people, obstacles, exits, and even the evacuation guide indicators are modeled as the original component of the PSO algorithm. It is convenient to setup the simulation environment upon the framework. Therefore, taking the proposed work as a research tool will be advantageous when the issue of evacuation simulation is investigated.

Luby transform (LT) codes implements an important property called ratelessness, meaning a fixed code rate is unnecessary and LT codes can complete the transmission without channel status. The property is advantageous to transmit over certain environments such as broadcasting in heterogeneous networks or transmitting data over unknown channels. For this reason, improving LT codes is a crucial research issue in recent years. The performance of LT codes is decided by the code length and a probability mass function, called degree distribution, used in the encoding process. To improve the performance of LT codes, many studies proposed to optimize the degree distribution by using methods in evolutionary computation. One of the key steps in the evolutionary process is to evaluate decision variables for comparing the fitness of each individual. In the optimization of LT codes, it needs to repeatedly simulate the encoding/decoding process with a given distribution and evaluate the performance over a sufficient number of runs. Hence, a lot of computational resource is necessary for the optimization of LT codes. In this paper, we propose a heuristic function to evaluate the performance of LT codes. The evaluation function estimates the expected fraction of unsolved symbols with the specified code length, reception overhead, and degree distribution. Based on the proposed function, a huge number of evaluations is possible for searching for better degree distributions. We first verify the practicality of the proposed function and then employ it in a multi-objective evolutionary algorithm to investigate the tradeoff of LT codes between the computational cost and decoding performance.

The synergy between exploration and exploitation has been a prominent issue in optimization. The rise of memetic algorithms, a category of optimization techniques which feature the explicit exploration-exploitation coordination, much accentuates this issue. While memetic algorithms have achieved remarkable success in a wide range of real-world applications, the key to a successful exploration-exploitation synergy still remains obscure. Manifold empirical results and theoretical derivations have been proposed and provided various perspectives from different algorithm-problem complexes to this issue. In our previous work, the concept of local search zones was proposed to provide an alternative perspective depicting the general behavior of memetic algorithms on a broad range of problems. In this work, based on the local search zone concept, we further investigate how the problem landscape and the way the algorithm explores and exploits the search space affect the performance of a memetic algorithm. The collaborative behavior of several representative archetypes of memetic algorithms, which exhibit different degrees of explorability and exploitability, are illustrated empirically and analytically on problems with different landscapes. As the empirical results consist with the local search zone concept and describe the behavior of various memetic algorithms on different problems, this work may reveal some essential design principals for memetic algorithms.

Luby Transform (LT) codes are a new member in the family of forward error correction codes without a fixed code rate. The property called rateless is attractive to researchers in last decade, and lots of studies have been proposed and attempted to improve the performance of LT codes. One variation is the use of a sparse degree distribution instead of a full one referred to in the encoding process of LT codes to reduce the search space. Observing a fact that the ability of a sparse degree distribution is limited by the nonempty degrees, we introduce a tag selection scheme to choose reasonable sparse degrees for LT codes in this paper. We firstly investigate the influence of different degrees on the error rate of LT codes and then propose a general selection algorithm based on our observations. After that, the covariance matrix adaptation evolution strategy (CMA-ES) is applied to find the optimal sparse degree distributions of which the degrees are defined by our selection algorithm. Finally, the experimental results are presented as evidence to show the proposed scheme is effective and practical.

The synergy between exploration and exploitation has been a prominent issue in optimization. The rise of memetic algorithms, a category of optimization techniques which feature the explicit exploration-exploitation coordination, much accentuates this issue. While memetic algorithms have achieved remarkable success in a wide range of real-world applications, the key to successful exploration-exploitation synergies still remains obscure as conclusions drawn from empirical results or theoretical derivations are usually quite algorithm specific and/or problem dependent. This paper aims to provide a theoretical model that can depict the collaboration between global search and local search in memetic computation on a broad class of objective functions. In the proposed model, the interaction between global search and local search creates a set of local search zones, in which the global optimal points reside, within the search space. Based on such a concept, the quasi-basin class (QBC) which categorizes problems according to the distribution of their local search zones is adopted. The subthreshold seeker, taken as a representative archetype of memetic algorithms, is analyzed on various QBCs to develop a general model for memetic algorithms. As the proposed model not only well describes the expected time for a simple memetic algorithm to find the optimal point on different QBCs but also consists with the observations made in previous studies in the literature, the proposed model may reveal important insights to the design of memetic algorithms in general.

Luby Transform code (LT code) has been a popular and practical technique in the field of channel coding since its proposal. One of the key components of LT code is a degree distribution which is used to determine the relationship between source data and codewords. Luby in his proposal suggested two general methods to construct feasible degree distributions. Such general designs work appropriately in typical situations but not optimally in most cases. To explore the full potential of LT code, in this work, we make the first attempt to introduce evolutionary algorithms to optimize the degree distribution in LT code. Degree distributions are encoded as real-valued vectors and evaluated by numerical simulation of LT code. For applications of different natures, two objectives are implemented to search good degree distributions with different decoding behavior. Compared with the original design, the experimental results are quite promising and demonstrate that the degree distribution can be customized for different purposes. In addition to manually adjusting the degree distribution as the common practice, the work presented in this paper provides an efficient alternative approach to use and adapt LT code for both practitioners and researchers.

Luby Transform code (LT code) is the first practical digital fountain code and has been widely used as basic components in many communication applications. The coding behavior of LT code is mainly decided by a probability distribution of codeword degrees. In order to customize a degree distribution for different purposes, multi-objective evolutionary algorithm is introduced to optimize degree distributions in this paper. Two critical performance indicators of LT code are considered in our experiments. Some applications hope to minimize the overhead of extra packets and some require to limit the computational cost of the coding system. To handle this problem, MOEA/D is applied to optimize two objectives simultaneously. We expect to obtain the Pareto front (PF) formed by partial optimal solutions and provide those available degree distributions to different LT code applications. Not only promising results are represented in this paper but also the behavior of LT code is thoroughly explored by optimizing the degree distribution according to multi-objectives.

An adaptive discretization method, called split-on-demand (SoD), enables estimation of distribution algorithms (EDAs) for discrete variables to solve continuous optimization problems. SoD randomly splits a continuous interval if the number of search points within the interval exceeds a threshold, which is decreased at every iteration. After the split operation, the nonempty intervals are assigned integer codes, and the search points are discretized accordingly. As an example of using SoD with EDAs, the integration of SoD and the extended compact genetic algorithm (ECGA) is presented and numerically examined. In this integration, we adopt a local search mechanism as an optional component of our back end optimization engine. As a result, the proposed framework can be considered as a memetic algorithm, and SoD can potentially be applied to other memetic algorithms. The numerical experiments consist of two parts: (1) a set of benchmark functions on which ECGA with SoD and ECGA with two well-known discretization methods: the fixed-height histogram (FHH) and the fixed-width histogram (FWH) are compared; (2) a real-world application, the economic dispatch problem, on which ECGA with SoD is compared to other methods. The experimental results indicate that SoD is a better discretization method to work with ECGA. Moreover, ECGA with SoD works quite well on the economic dispatch problem and delivers solutions better than the best known results obtained by other methods in existence.

The probabilistic model building performed by estimation of distribution algorithms (EDAs) enables these methods to use advanced techniques of statistics and machine learning for automatic discovery of problem structures. However, in some situations, it may not be possible to completely and accurately identify the whole problem structure by probabilistic modeling due to certain inherent properties of the given problem. In this work, we illustrate one possible cause of such situations with problems consisting of structures with unequal fitness contributions. Based on the illustrative example, we introduce a notion that the estimated probabilistic models should be inspected to reveal the effective search directions and further propose a general approach which utilizes a reserved set of solutions to examine the built model for likely inaccurate fragments. Furthermore, the proposed approach is implemented on the extended compact genetic algorithm (ECGA) and experiments are performed on several sets of additively separable problems with different scaling setups. The results indicate that the proposed method can significantly assist ECGA to handle problems comprising structures of disparate fitness contributions and therefore may potentially help EDAs in general to overcome those situations in which the entire problem structure cannot be recognized properly due to the temporal delay of emergence of some promising partial solutions.

Multi-objective optimization is an essential and challenging topic in the domains of engineering and computation because real-world problems usually include several conflicting objectives. Current trends in the research of solving multi-objective problems (MOPs) require that the adopted optimization method provides an approximation of the Pareto set such that the user can understand the tradeoff between objectives and therefore make the final decision. Recently, an efficient framework, called MOEA/D, combining decomposition techniques in mathematics and optimization methods in evolutionary computation was proposed. MOEA/D decomposes a MOP to a set of single-objective problems (SOPs) with neighborhood relationship and approximates the Pareto set by solving these SOPs. In this paper, we attempt to enhance MOEA/D by proposing two mechanisms. To fully employ the information obtained from neighbors, we introduce a guided mutation operator to replace the differential evolution operator. Moreover, a update mechanism utilizing a priority queue is proposed for performance improvement when the SOPs obtained by decomposition are not uniformly distributed on the Pareto font. Different combinations of these approaches are compared based on the test problem instances proposed for the CEC 2009 competition. The set of problem instances include unconstrained and constrained MOPs with variable linkages. Experimental results are presented in the paper, and observations and discussion are also provided.

Genetic algorithms and the descendant methods have been deemed robust and practical. To enhance the capabilities of genetic algorithms, tremendous effort has been invested in the field of evolutionary computation. One of the major trends to enhance genetic algorithms is to extract and exploit the relationship among variables, such as estimation of distribution algorithms and perturbation-based methods. In this study, we make an attempt to enable inductive linkage identification (ILI) to detect general problem structures, in which one variable may link to an arbitrary number of other variables. Our results indicate that the proposed technique can successfully detect the given problem structure.

Estimation of distribution algorithms (EDAs) are a class of evolutionary algorithms that capture the likely structure of promising solutions by explicitly building a probabilistic model and utilize the built model to guide the further search. It is presumed that EDAs can detect the structure of the problem by recognizing the regularities of the promising solutions. However, in certain situations, EDAs are unable to discover the entire structure of the problem because the set of promising solutions on which the model is built contains insufficient information regrading some parts of the problem and renders EDAs incapable of processing those parts accurately. In this work, we firstly propose a general concept that the estimated probabilistic models should be inspected to reveal the effective search directions. Based on that concept, we design a practical approach which utilizes a reserved set of solutions to examine the built model for the fragments that may be inconsistent with the actual problem structure. Furthermore, we provide an implementation of the designed approach on the extended compact genetic algorithm (ECGA) and conduct numerical experiments. The experimental results indicate that the proposed method can significantly assist ECGA to handle problems comprising building blocks of disparate scalings.

This paper extends an adaptive discretization method, Split-on-Demand (SoD), to be capable of handling multidimensional continuous search spaces. The proposed extension is called multidimensional Split-on-Demand (mSoD), which considers multiple dimensions of the search space as a whole instead of independently discretizing each dimension as SoD does. In this study, we integrate mSoD and SoD with the extended compact genetic algorithm (ECGA) to numerically examine the effectiveness and performance of mSoD and SoD on the problems with and without linkage among dimensions of the search space. The experimental results indicate that mSoD outperforms SoD on both types of the test problems and that mSoD can offer better scalability, stability, and accuracy. The behavior of mSoD is discussed, followed by the potential future work.

The purpose of linkage identification in genetic and evolutionary algorithms is to detect the strongly related variables of the fitness function. If such linkage information can be acquired, the crossover or recombination operator can accordingly mix the discovered sub-solutions effectively without disrupting them. In this paper, we propose a new linkage identification technique, called inductive linkage identification (ILI), employing perturbation with decision tree induction. With the proposed scheme, the linkage information can be obtained by first constructing an ID3 decision tree to learn the mapping from the population of solutions to their corresponding fitness differences caused by perturbations and then inspecting the constructed decision tree for variables exhibiting strong interdependencies with one another. The numerical results show that the proposed technique can accomplish the identical linkage identification task with a lower number of function evaluations compared to similar methods proposed in the literature. Moreover, the proposed technique is also shown being able to handle both uniformly scaled and exponentially scaled problems.

This paper presents a uniform conceptual model based on the particle swarm optimization (PSO) paradigm to simulate crowds in computer graphics. According to the mechanisms of PSO, each person (particle) in the crowd (swarm) can adopt the information to search a path from the initial position to the specified target (optimum) automatically. However, PSO aims to obtain the optimal solution, while the purpose of this study concentrates on the generated paths of particles. Hence, in order to generate appropriate paths of people in a crowd, we propose a method to employ the computational facilities provided in PSO. The proposed model is simple, uniform, and easy to implement. The results of simulations demonstrate that using PSO with the proposed technique can generate appropriate non deterministic, non-colliding paths in several different scenarios, including static obstacles, moving targets, and multiple crowds.

In this paper, we introduce fault tolerance to XCS and propose a new XCS framework called XCS with Fault Tolerance (XCS/FT). As an important branch of learning classifier systems, XCS has been proven capable of evolving maximally accurate, maximally general problem solutions. However, in practice, it oftentimes generates a lot of rules, which lower the readability of the evolved classification model, and thus, people may not be able to get the desired knowledge or useful information out of the model. Inspired by the fault tolerance mechanism proposed in field of data mining, we devise a new XCS framework by integrating the concept and mechanism of fault tolerance into XCS in order to reduce the number of classification rules and therefore to improve the readability of the generated prediction model. A series of $N$-multiplexer experiments, including 6-bit, 11-bit, 20-bit, and 37-bit multiplexers, are conducted to examine whether XCS/FT can accomplish its goal of design. According to the experimental results, XCS/FT can offer the same level of prediction accuracy on the test problems as XCS can, while the prediction model evolved by XCS/FT consists of significantly fewer classification rules.

In this paper, we propose a new approach that consists of the extended compact genetic algorithm (ECGA) and split-on-demand (SoD), an adaptive discretization technique, to economic dispatch (ED) problems with nonsmooth cost functions. ECGA is designed for handling problems with decision variables of the discrete type, while the decision variables of ED problems are oftentimes real numbers. Thus, in order to employ ECGA to tackle ED problems, SoD is utilized for discretizing the continuous decision variables and works as the interface between ECGA and the ED problem. Furthermore, ED problems in practice are usually hard for traditional mathematical programming methodologies because of the equality and inequality constraints. Hence, in addition to integrating ECGA and SoD, in this study, we devise a repair operator specifically for making the infeasible solutions to satisfy the equality constraint. To examine the performance and effectiveness, we apply the proposed framework to two different-sized ED problems with nonsmooth cost function considering the valve-point effects. The experimental results are compared to those obtained by various evolutionary algorithms and demonstrate that handling ED problems with the proposed framework is a promising research direction.

Evolution strategy (ES) and particle swarm optimization (PSO) are two of the most popular research topics for tackling real-parameter optimization problems in evolutionary computation. Both of them have strengths and weaknesses for their different search behaviors and methodologies. In ES, mutation, as the main operator, tries to find good solutions around each individual. While in PSO, particles are moving toward directions determined by certain global information, such as the global best particle. In order to leverage the specialties offered by both sides to our advantage, this paper combines the essential mechanism of ES and the key concept of PSO to develop a new hybrid optimization methodology, called particle swarm guided evolution strategy. We introduce swarm intelligence to the ES mutation framework to create a new mutation operator, called guided mutation, and integrate the guided mutation operator into ES. Numerical experiments are conducted on a set of benchmark functions, and the experimental results indicate that PSGES is a promising optimization methodology as well as an interesting research direction.

In this paper, we develop a new optimization framework that consists of the extended compact genetic algorithm (ECGA) and split-on-demand (SoD), an adaptive discretization technique, to tackle the characteristic determination problem for solid state devices. As most decision variables of characteristic determination problems are real numbers due to the modeling of physical phenomena, and ECGA is designed for handling discrete-type problems, a specific mechanism to transform the variable types of the two ends is in order. In the proposed framework, ECGA is used as a back-end optimization engine, and SoD is adopted as the interface between the engine and the problem. Moreover, instead of one mathematical model with various parameters, characteristic determination is in fact a set of problems of which the mathematical formulations may be very different. Therefore, in this study, we employ the proposed framework on three study cases to demonstrate that the technique proposed in the domain of evolutionary computation can provide not only the high quality optimization results but also the flexibility to handle problems of different formulations.

In this paper, we try to improve the performance of the particle swarm optimizer by incorporating the linkage concept, which is an essential mechanism in genetic algorithms, and design a new linkage identification technique called dynamic linkage discovery to address the linkage problem in real-parameter optimization problems. Dynamic linkage discovery is a costless and effective linkage recognition technique that adapts the linkage configuration by employing only the selection operator without extra judging criteria irrelevant to the objective function. Moreover, a recombination operator that utilizes the discovered linkage configuration to promote the cooperation of particle swarm optimizer and dynamic linkage discovery is accordingly developed. By integrating the particle swarm optimizer, dynamic linkage discovery, and recombination operator, we propose a new hybridization of optimization methodologies called particle swarm optimization with recombination and dynamic linkage discovery (PSO-RDL). In order to study the capability of PSO-RDL, numerical experiments were conducted on a set of benchmark functions as well as on an important real-world application. The benchmark functions used in this paper were proposed in the 2005 Institute of Electrical and Electronics Engineers Congress on Evolutionary Computation. The experimental results on the benchmark functions indicate that PSO-RDL can provide a level of performance comparable to that given by other advanced optimization techniques. In addition to the benchmark, PSO-RDL was also used to solve the economic dispatch (ED) problem for power systems, which is a real-world problem and highly constrained. The results indicate that PSO-RDL can successfully solve the ED problem for the three-unit power system and obtain the currently known best solution for the 40-unit system.

In the realm of data mining, several key issues exists in the traditional classification algorithms, such as low readability, large rule number, and low accuracy with information losing. In this paper, we propose a new classification methodology, called fault tolerance XCS in integer (FTXI), by extending XCS to handle conditions in integers and integrating the mechanism of fault tolerance in the context of data mining into the framework of XCS. We also design and generate appropriate artificial data sets for examining and verifying the proposed method. Our experiments indicate that FTXI can provide the least rule number, obtain high prediction accuracy, and offer rule readability, compared to C4.5 and XCS in integer without fault tolerance.

This paper proposes an adaptive discretization method, called Split-on-Demand (SoD), to enable the probabilistic model building genetic algorithm (PMBGA) to solve optimization problems in the continuous domain. The procedure, effect, and usage of SoD are described in detail. As an example, the integration of SoD and the extended compact genetic algorithm (ECGA), named real-coded ECGA (rECGA), is presented and numerically examined. The experimental results indicate that rECGA works well and SoD is effective. The behavior of SoD is analyzed and discussed, followed by the potential future work for SoD.

This paper proposes the CFE framework---Composition, Feedback, and Evolution---and presents an interactive music composition system. The system composes short, manageable pieces of music by interacting with users. The most important features of the system include creating customized music according to the user preference and providing the facilities specifically designed for producing large amounts of music. We present the structure as well as the implementation of the system and the auxiliary functionalities that enhance the system. We also introduce the auto-feedback test with which we verify and evaluate the interactive music composition system.

Extended compact genetic algorithm (ECGA) is an algorithm that can solve hard problems in the binary domain. ECGA is reliable and accurate because of the capability of detecting building blocks, but certain difficulties are encountered when we directly apply ECGA to problems in the integer domain. In this paper, we propose a new algorithm that extends ECGA, called integer extended compact genetic algorithm (iECGA). iECGA uses a modified probability model and inherits the capability of detecting building blocks from ECGA. iECGA is specifically designed for problems in the integer domain and can avoid the difficulties that ECGA encounters. With the experimental results, we show the performance comparisons between ECGA, iECGA, and a simple GA. The results indicate that iECGA has good performance on problems in the integer domain.

In this paper, we introduce the recombination operator with the technique of dynamic linkage discovery to particle swarm optimization (PSO) in order to improve the performance of PSO. Numerical experiments are conducted on a set of carefully designed benchmark functions and demonstrate good performance achieved by the proposed methodology.

This paper identifies the sequential behavior of the linkage learning genetic algorithm, introduces the tightness time model for a single building block, and develops the connection between the sequential behavior and the tightness time model. By integrating the first-building-block model based on the sequential behavior, the tightness time model, and the connection between these two models, a convergence time model is constructed and empirically verified. The proposed convergence time model explains the exponentially growing time required by the linkage learning genetic algorithm when solving uniformly scaled problems.

This paper identifies the sequential behavior of the linkage learning genetic algorithm (LLGA), introduces the tightness time model for a single building block, and develops the connection between sequential behavior and the tightness time model. By integrating the first building-block model based on sequential behavior, the tightness time model, and the connection between these two models, a convergence time model is then constructed and empirically verified. The proposed convergence time model explains the exponentially growing time required by LLGA when solving uniformly scaled problems.