Optimization and FROC analysis of rule-based detection schemes using a multiobjective approach

Computerized detection schemes have the potential of increasing diagnostic accuracy in medical imaging by alerting radiologists to lesions that they initially overlooked. These schemes typically employ multiple parameters such as threshold values or filter weights to arrive at a detection decision. In order for the system to have high performance, the values of these parameters need to be set optimally. Conventional optimization techniques are designed to optimize a scalar objective function. The task of optimizing the performance of a computerized detection scheme, however, is clearly a multiobjective problem: we wish to simultaneously improve the sensitivity and false-positive rate of the system. In this work we investigate a multiobjective approach to optimizing computerized rule-based detection schemes. In a multiobjective optimization, multiple objectives are simultaneously optimized, with the objective now being a vector-valued function. The multiobjective optimization problem admits a set of solutions, known as the Pareto-optimal set, which are equivalent in the absence of any information regarding the preferences of the objectives. The performances of the Pareto-optimal solutions can be interpreted as operating points on an optimal free-response receiver operating characteristic (FROC) curve, greater than or equal to the points on any possible FROC curve for a given dataset and detection scheme. It is demonstrated that generating FROC curves in this manner eliminates several known problems with conventional FROC curve generation techniques for rule-based detection schemes. We employ the multiobjective approach to optimize a rule-based scheme for clustered microcalcification detection that has been developed in our laboratory.     

Multiobjective gas turbine engine controller design using geneticalgorithms

This paper describes the use of multiobjective genetic algorithms (MOGAs) in the design of a multivariable control system for a gas turbine engine. The mechanisms employed to facilitate multiobjective search with the genetic algorithm are described with the aid of an example. It is shown that the MOGA confers a number of advantages over conventional multiobjective optimization methods by evolving a family of Pareto-optimal solutions rather than a single solution estimate. This allows the engineer to examine the trade-offs between the different design objectives and configurations during the course of an optimization. In addition, the paper demonstrates how the genetic algorithm can be used to search in both controller structure and parameter space thereby offering a potentially more general approach to optimization in controller design than traditional numerical methods. While the example in the paper deals with control system design, the approach described can be expected to be applicable to more general problems in the fields of computer aided design (CAD) and computer aided engineering (CAE)     

Survey of research on application of heuristic algorithm in machine learning

Aiming at the problems existing in the application of machine learning algorithm,an optimization system of the machine learning model based on the heuristic algorithm was constructed.Firstly,the existing types of heuristic algorithms and the modeling process of heuristic algorithms were introduced.Then,the advantages of the heuristic algorithm were illustrated from its applications in machine learning,including the parameter and structure optimization of neural network and other machine learning algorithms,feature optimization,ensemble pruning,prototype optimization,weighted voting ensemble and kernel function learning.Finally,the heuristic algorithms and their development directions in the field of machine learning were given according to the actual needs.     

Multiobjective genetic algorithm applied to aerodynamic design ofcascade airfoils

A multiobjective genetic algorithm (GA) based on Fonseca-Fleming's Pareto-based ranking and fitness-sharing techniques has been applied to aerodynamic shape optimization of cascade airfoil design. Airfoil performance is evaluated by a Navier-Stokes code. Evaluation of GA population is parallelized on the Numerical Wind Tunnel, a parallel vector machine. The present multiobjective design seeks high pressure rise, high flow turning angle, and low total pressure loss at a low Mach number. Pareto solutions that perform better than existing control diffusion airfoils were obtained     

Multiobjective genetic optimization of diagnostic classifiers withimplications for generating receiver operating characteristiccurves

It is well understood that binary classifiers have two implicit objective functions (sensitivity and specificity) describing their performance. Traditional methods of classifier training attempt to combine these two objective functions (or two analogous class performance measures) into one so that conventional scalar optimization techniques can be utilized. This involves incorporating a priori information into the aggregation method so that the resulting performance of the classifier is satisfactory for the task at hand. We have investigated the use of a niched Pareto multiobjective genetic algorithm (GA) for classifier optimization. With niched Pareto GA's, an objective vector is optimized instead of a scalar function, eliminating the need to aggregate classification objective functions. The niched Pareto GA returns a set of optimal solutions that are equivalent in the absence of any information regarding the preferences of the objectives. The a priori knowledge that was used for aggregating the objective functions in conventional classifier training can instead be applied post-optimization to select from one of the series of solutions returned from the multiobjective genetic optimization. We have applied this technique to train a linear classifier and an artificial neural network (ANN), using simulated datasets. The performances of the solutions returned from the multiobjective genetic optimization represent a series of optimal (sensitivity, specificity) pairs, which can be thought of as operating points on a receiver operating characteristic (ROC) curve. All possible ROC curves for a given dataset and classifier are less than or equal to the ROC curve generated by the niched Pareto genetic optimization.     

Evolutionary Multiobjective Optimization for the Pickup and Delivery Problem with Time Windows and Demands

This paper studies an evolutionary algorithm to solve a new multiobjective optimization problem, the Pickup and Delivery Problem with Time Windows and Demands (PDP-TW-D), which is applicable to operational optimization in various mobile network systems. With respect to multiple optimization objectives, PDP-TW-D is to find a set of Pareto-optimal routes for a fleet of vehicles (e.g., mobile robots, drones and autonomous heavy-haulage trucks) in order to serve given transportation requests. The proposed algorithm uses a population of individuals, each of which represents a solution candidate, and evolves them through generations to seek the Pareto-optimal solutions. In addition to the evolution-based global search process, the proposed algorithm allows individuals to improve their optimality in each generation with a local search process, which is designed based on iterative neighborhood search. Experimental results demonstrate that the integration of global and local search processes improves the optimality of individuals and expedites convergence speed. The proposed algorithm outperforms two well-known existing EMOAs, NSGA-II and MOEA/D, in relatively large-scale problems that have up to 400 pickup and delivery locations.     

Multiobjective optimization design of Yagi-Uda antenna

An optimization method, such as the steepest gradient methods, could not easily obtain globally optimum solutions for devising antenna design parameters that allow the antenna to simultaneously improve multiple performances such as gain, sidelobe level, and input impedance. The genetic algorithm (GA) is suitable for empirically solving optimization problems and is effective in designing an antenna. In particular, this method can solve the multiobjective optimization problem using various Pareto-optimal solutions in an extremely efficient manner. In this paper, the Pareto GA, by which various Pareto-optimal solutions for each objective function (performance) can be obtained in a single trial of a numerical simulation and which enables the selection of parameters in accordance with the design requirement, is applied to the multiobjective optimization design of the Yagi-Uda antenna. The effectiveness of the Pareto GA was demonstrated by comparing the performances obtained by the Pareto GA with those of the previously reported values, which were obtained by the conventional GA, and with the values of the design benchmark reference.     

Multiobjective placement of electronic components using evolutionary algorithms

The optimal placement of electronic components on a printed circuit board is a well-studied optimization task. However, despite the involvement of multiple conflicting objectives, researchers have mainly used a single objective of minimizing the overall wire length or minimizing the overall heat generation or minimizing the overall time delay in its functioning. In this paper, the problem is treated as a two-objective optimization problem of minimizing the overall wire length and minimizing the failure-rate of the board arising due to uneven local heat accumulation. The proposed strategy uses a novel representation procedure and a multiobjective evolutionary algorithm capable of finding multiple Pareto-optimal solutions simultaneously. Moreover, the flexibility and efficacy of the proposed strategy have been demonstrated by simultaneously optimizing the placement of components and the layout of the board. The convergence and the extent of spread obtained in the solutions reliably by repetitive applications of the proposed procedure should encourage further application of the approach to more complex placement design problems.     

Multiobjective programming using uniform design and genetic algorithm

The notion of Pareto-optimality is one of the major approaches to multiobjective programming. While it is desirable to find more Pareto-optimal solutions, it is also desirable to find the ones scattered uniformly over the Pareto frontier in order to provide a variety of compromise solutions to the decision maker. We design a genetic algorithm for this purpose. We compose multiple fitness functions to guide the search, where each fitness function is equal to a weighted sum of the normalized objective functions and we apply an experimental design method called uniform design to select the weights. As a result, the search directions guided by these fitness functions are scattered uniformly toward the Pareto frontier in the objective space. With multiple fitness functions, we design a selection scheme to maintain a good and diverse population. In addition, we apply the uniform design to generate a good initial population and design a new crossover operator for searching the Pareto-optimal solutions. The numerical results demonstrate that the proposed algorithm can find the Pareto-optimal solutions scattered uniformly over the Pareto frontier.     

Hyperparameter Optimization for Machine Learning Models Based on Bayesian Optimization

Hyperparameters are important for machine learning algorithms since they directly control the behaviors of training algorithms and have a significant effect on the performance of machine learning models. Several techniques have been developed and successfully applied for certain application domains. However, this work demands professional knowledge and expert experience. And sometimes it has to resort to the brute-force search. Therefore, if an efficient hyperparameter optimization algorithm can be developed to optimize any given machine learning method, it will greatly improve the efficiency of machine learning. In this paper, we consider building the relationship between the performance of the machine learning models and their hyperparameters by Gaussian processes. In this way, the hyperparameter tuning problem can be abstracted as an optimization problem and Bayesian optimization is used to solve the problem. Bayesian optimization is based on the Bayesian theorem. It sets a prior over the optimization function and gathers the information from the previous sample to update the posterior of the optimization function. A utility function selects the next sample point to maximize the optimization function. Several experiments were conducted on standard test datasets. Experiment results show that the proposed method can find the best hyperparameters for the widely used machine learning models, such as the random forest algorithm and the neural networks, even multi-grained cascade forest under the consideration of time cost.     

Cooperative spectrum sensing in cognitive radio network using multiobjective evolutionary algorithms and fuzzy decision making

The cognitive radio has emerged as a potential solution to the problem of spectrum scarcity. Spectrum sensing unit in cognitive radio deals with the reliable detection of primary user’s signal. Cooperative spectrum sensing exploits the spatial diversity between cognitive radios to improve sensing accuracy. The selection of the weight assigned to each cognitive radio and the global decision threshold can be formulated as a constrained multiobjective optimization problem where probabilities of false alarm and detection are the two conflicting objectives. This paper uses evolutionary algorithms to solve this optimization problem in a multiobjective framework. The simulation results offered by different algorithms are assessed and compared using three performance metrics. This study shows that our approach which is based on the concept of cat swarm optimization outperforms other algorithms in terms of quality of nondominating solutions and efficient computation. A fuzzy logic based strategy is used to find out a compromise solution from the set of nondominated solutions. Different tests are carried out to assess the stability of the simulation results offered by the heuristic evolutionary algorithms. Finally the sensitivity analysis of different parameters is performed to demonstrate their impact on the overall performance of the system.     

Designing resilient networks using multicriteria metaheuristics

The paper deals with the design of resilient networks that are fault tolerant against link failures. Usually, fault tolerance is achieved by providing backup paths, which are used in case of an edge failure on a primary path. We consider this task as a multiobjective optimization problem: to provide resilience in networks while minimizing the cost subject to capacity constraint. We propose a stochastic approach, which can generate multiple Pareto solutions in a single run. The feasibility of the proposed method is illustrated by considering several network design problems using a single weighted average of objectives and a direct multiobjective optimization approach using the Pareto dominance concept.     

Evolutionary algorithms for multiobjective and multimodal optimization of diagnostic schemes

This paper addresses the optimization of noninvasive diagnostic schemes using evolutionary algorithms in medical applications based on the interpretation of biosignals. A general diagnostic methodology using a set of definable characteristics extracted from the biosignal source followed by the specific diagnostic scheme is presented. In this framework, multiobjective evolutionary algorithms are used to meet not only classification accuracy but also other objectives of medical interest, which can be conflicting. Furthermore, the use of both multimodal and multiobjective evolutionary optimization algorithms provides the medical specialist with different alternatives for configuring the diagnostic scheme. Some application examples of this methodology are described in the diagnosis of a specific cardiac disorder-paroxysmal atrial fibrillation.     

Multiobjective GA optimization using reduced models

In this paper, we propose a novel method for solving multiobjective optimization problems using reduced models. Our method, called objective exchange genetic algorithm for design optimization (OEGADO), is intended for solving real-world application problems. For such problems, the number of objective evaluations performed is a critical factor as a single objective evaluation can be quite expensive. The aim of our research is to reduce the number of objective evaluations needed to find a well-distributed sampling of the Pareto-optimal region by applying reduced models to steady-state multiobjective GAs. OEGADO runs several GAs concurrently with each GA optimizing one objective and forming a reduced model of its objective. At regular intervals, each GA exchanges its reduced model with the others. The GAs use these reduced models to bias their search toward compromise solutions. Empirical results in several engineering and benchmark domains comparing OEGADO with two state-of-the-art multiobjective evolutionary algorithms show that OEGADO outperformed them for difficult problems.     

Online multiobjective Pareto optimal dynamic scheduling of semiconductor back-end using conjunctive simulated scheduling

A discrete event simulation based "online near-real-time" dynamic multiobjective scheduling system has been conceptualized, designed, and developed to achieve Pareto optimal solutions in a complex manufacturing environment of semiconductor back-end. Our approach includes the use of linear weighted aggregation optimization approach for multiple objectives and auto simulation model generation for online simulation. Developed concepts are implemented at a semiconductor back-end site and are in use. The impact of the system includes a better customer delivery achievement, consistent cycle time with narrower distribution, improved machine utilization, reduction in the time that planners and manufacturing personnel spend on scheduling, and more predictable and repeatable manufacturing performance. In addition, it enables managers and senior planners to carry out "what now" analysis to make effective current decisions and "what if" analysis to plan for the future.     

Virtual network function–forwarding graph embedding: A genetic algorithm approach

Network function virtualization (NFV) provides a simple and effective mean to deploy and manage network and telecommunications' services. A typical service can be expressed in the form of a virtual network function–forwarding graph (VNF‐FG). Allocating a VNF‐FG is equivalent to place VNFs and virtual links onto a given substrate network considering resources and quality‐of‐service (QoS) constraints. The deployment of VNF‐FGs in large‐scale networks, such that QoS measures and deployment cost are optimized, is an emerging challenge. Single‐objective VNF‐FGs allocation has been addressed in existing literature; however, there is still a lack of studies considering multiobjective VNF‐FGs allocation. In addition, it is not trivial to obtain optimal VNF‐FGs allocation due to its high computational complexity even in case of single‐objective VNF‐FGs allocation. Genetic algorithms (GAs) have been proved its ability in coping with multiobjective optimization problems; thus, we propose a GA‐based scheme to solve multiobjective VNF‐FGs allocation problem in this paper. The numerical results confirm that the proposed scheme can provide near Pareto‐optimal solutions within a short execution time.     

基于多目标进化策略算法的DNA核酸编码设计

设计高质量的核酸分子集合能有效提高DNA计算的可靠性、有效性和可求解问题的规模。DNA分子需要满足热力学约束、相似度约束、GC含量约束等多个相互冲突的目标函数,是典型的多目标优化问题。该文提出一种多目标进化策略(MOES)算法求解DNA分子序列设计问题,算法设计了随机碱基变异算子实现高效的局部搜索和全局搜索。改进的评价函数综合考虑了候选解的支配关系和冲突目标的平衡程度,选取符合DNA编码约束的核酸序列。实验结果证明,该文提出的算法具有高效的搜索效率和快速收敛能力,可以产生高质量的DNA序列集合,优于其他对比算法产生的DNA分子序列集合。     

Clustering Methods for Agent Distribution Optimization

Multiagent systems consist of a collection of agents that directly interact usually via a form of message passing. Information about these interactions can be analyzed in an online or offline way to identify clusters of agents that are related. The first part of this paper is dedicated to a formal definition of a proposed dynamic model for agent clustering and experimental results that demonstrate applicability of this novel approach. The main contribution is the ability to discover and visualize communication neighborhoods of agents at runtime, which is a novel approach not attempted so far. The second part of this paper deals with a static agent clustering problem where equally sized clusters with maximal intracluster communication among agents are sought in order to efficiently distribute agents across multiple execution units. The weakness of standard clustering approaches for solving this type of clustering problem is shown. First, these algorithms optimize the generated clustering with respect to just one criterion, and therefore, yield solutions with inferior quality relative to the other criteria. Second, the algorithms are deterministic; thus they can produce just a single solution for the given data. A multiobjective clustering approach based on an iterative optimization evolutionary algorithm called multiobjective prototype optimization with evolved improvement steps (mPOEMS) is proposed and its advantages are demonstrated. The most important observation is that mPOEMS produces numerous high-quality solutions in a single run from which a user can choose the best one. The best solutions found by mPOEMS are significantly better than the solutions generated by the compared clustering algorithms.      

Battery-Aware Adaptive Modulation With QoS Constraints

We consider the problem of joint battery management and modulation optimization for maximizing the lifetime of a battery-powered autonomic wireless system while satisfying certain quality-of-service constraints. Our objective is to design the optimal transmission schemes adapting to the battery state, the delay constraints, and channel conditions. A delay-constrained data-transmission scenario is considered, where the battery state information is only available at the beginning of the entire transmission duration. The corresponding problem is formulated as a multiobjective stochastic optimization problem, which is hard to solve. We approximately transform the original problem to a multiobjective deterministic optimization problem by using a lower bound to replace the objective function of the original problem. The optimal, as well as efficient suboptimal solutions are developed. Simulation results are provided to demonstrate that the proposed algorithms can significantly increase the lifetime of the battery-powered autonomic wireless systems.     

A Jumping Genes Paradigm: Theory, Verification and Applications

A new evolutionary computing algorithm on the basis of "jumping genes" phenomenon is presented in this article. It emulates the gene transposition in the genome that was discovered by N obel Laureate D r. Barbara M cClintock from her work on maize chromosome. T he principle of jumping genes, adopted for evolutionary computing, is outlined and the procedures for executing the computational optimization are provided. M athematical derivation of the S chema Theorem is briefly discussed, which is established to demonstrate the searching capacity of the newly proposed algorithm, in terms of convergence and diversity. T he algorithm is found to be robust and provides outcomes in speed and accuracy, while the solutions are widely spread along the Pareto-optimal front when a multiobjective problem is tackled. T o further reinforce the jumping genes proposition, some typical engineering design problems are included. T he obtained results have indicated that this new algorithm is indeed capable of searching multiobjective solutions including the extreme solutions at both ends of the Pareto-optimal front.