Search

This work describes a new surrogate-assisted constraint-handling technique (CHT) for parametric multi-objective evolutionary algorithms, called Bayesian CHT. Parametric optimization finds optimal solutions as a function of one or more exogenous variables. The solution is a family of Pareto frontiers called the parameterized Pareto frontier (PPF). CHTs from non-parametric multi-objective evolutionary algorithms do not produce a good sampling of solutions on the PPF. The proposed CHT addresses this using Bayesian methods. A Gaussian process classifier serves as an uncertainty-quantified surrogate for problem constraints, enabling active learning and a novel repair mechanism that promotes sampling along the PPF. The new technique is evaluated on a suite of 36 test problems and two engineering cases of structural design. Quantitative results show that the proposed Bayesian CHT outperforms several state-of-the-art algorithms in most cases. From a qualitative perspective, the nondominated solutions are visualized to support the superiority of the new approach, which achieves a better spread of solutions on the PPF. The results of engineering studies also indicate that the new approach is computationally more efficient than the others. [ABSTRACT FROM AUTHOR]

This paper goal is to provide a more effective algorithm for reducing the amount of test cases in a test suit during the regression testing phase. This algorithm divides the entire test suit into equivalence classes at first then apply boundary value coverage to select test case out of repeated test cases with same importance in suit. This algorithm is based on the concept that before selecting best test cases out of repeated test case in test suit to prepare reduced test suit we can divide all test cases in number of equivalence classes to reduce test cases by great extent. This paper proposed a method of experimentation involving test cases from different software application areas. Furthermore, we discuss a case study to verify and check new algorithm for its efficiency, for that we apply our algorithm on one of the program or group of programs. This complete proposed methodology shall be applied to different software applications belonging to soft computing, engineering software, and financial software. This wok has not been done earlier as found in the literature survey performed by us. Minimization in test cases would lead to lesser testing effort and desirable test completion. [ABSTRACT FROM AUTHOR]

In recent decades, multi-objective evolutionary algorithms (MOEAs) have been evaluated on artificial test problems with unrealistic characteristics, leading to uncertain conclusions about their efficacy in real-world applications. To address this issue, a few benchmark test suites comprising real-world problems have been proposed for MOEAs, encompassing numerous multi-objective problems and a select few many-objective problems. Given the distinct challenges posed by many-objective optimization problems (MaOPs) and their inherent difficulty, developing a test suite that includes real-world problems with many conflicting objectives is crucial. Hence, in this paper, we propose a comprehensive test suite for benchmarking the many-objective real-world complex problems. This test suite consists of 11 problems collected from different disciplines of engineering. Furthermore, we comprehensively analyzed the problems in our newly proposed test suite, employing eight state-of-the-art algorithms rooted in various fundamental principles specifically designed to address MaOPs. The experimental findings highlight the strong performance of indicator-based, weight-vector-based decomposition, Pareto-dominance-based, and hybrid MOEAs on the proposed test suite. In contrast, reference-vector-based decomposition approaches, Pareto front shape estimation-based methods, and multi-evolution approaches exhibit relatively weaker performance.

There are several areas of knowledge in which (pseudo-)random numbers are necessary, for example, in statistical–mathematical simulation or in cryptography and system security, among others. Depending on the area of application, it will be necessary that the sequences used meet certain requirements. In general, randomness and uniformity conditions are required in the generated sequences, which are checked with statistical tests, and conditions on sequence unpredictability if the application is in security. In the present work, a literature review on cryptographically secure pseudo-random number generators (CSPRNGs) is carried out, they are implemented, and a critical analysis of their statistical quality and computational efficiency is performed. For this purpose, different programming languages will be used, and the sequences obtained will be checked by means of the NIST Statistical Test Suite (NIST STS). In addition, a user's guide will be provided to allow the selection of one generator over another according to its statistical properties and computational implementation characteristics. [ABSTRACT FROM AUTHOR]

In software engineering, testing has long been a research area of software maintenance. Testing is extremely expensive, and there is no guarantee that all defects will be found within a single round of testing. Therefore, fixing defects that are not discovered by a single round of testing is important for reducing the test costs. During the software maintenance process, testing is conducted within the scope of a set of test cases called a test suite. Mutation testing is a method that uses mutants to evaluate whether the test cases of the test suite are appropriate. In this paper, an approach is proposed that uses the mutants of a mutation test to identify defects that are not discovered through a single round of testing. The proposed method simultaneously applies two or more mutants to a single program to define and record the relationships between different lines of code. In turn, these relationships are examined using the defects that were discovered by a single round of testing, and possible defects are recommended from among the recorded candidates. To evaluate the proposed method, a comparative study was conducted using the fault localization method, which is commonly employed in defect prediction, as well as the Defects4J defect prediction dataset, which is widely used in software defect prediction. The results of the evaluation showed that the proposed method achieves a better performance than seven other fault localization methods (Tarantula, Ochiai, Opt2, Barinel, Dstar2, Muse, and Jaccard).

State-based testing is a challenging area in the software testing field. This paper proposes a set of algorithms to generate test cases from a state chart diagram based on various coverage criteria. The objective is to find various types of state-based faults by covering states and transitions of an object. First, the state chart diagram is transformed into an intermediate graph, State Chart Intermediate Graph (SCIG). Then, for a given coverage criteria, traversing method is applied on SCIG to generate test cases. Different trees from SCIG based on various coverage criteria are extracted for test case generation. Various coverage criteria such as All Transition (AT), Round Trip Path (RTP) and All Transition Pair (ATP) are considered. We introduce algorithms for two most efficient state-based criteria, RTP and ATP. Two case studies, Stack Operation and Vending Machine Automation system, are discussed throughout the paper. We experimentally observed that (i) AT consumes the most test resources (ii) ATP can’t achieve 100% transition coverage (iii) test cases generated based on RTP is efficient, and it overcomes the transition explosion problem of AT. This analysis is beneficial in the area of semi-automatic test case generation in model-based testing.

There are several areas of knowledge in which (pseudo-)random numbers are necessary, for example, in statistical–mathematical simulation or in cryptography and system security, among others. Depending on the area of application, it will be necessary that the sequences used meet certain requirements. In general, randomness and uniformity conditions are required in the generated sequences, which are checked with statistical tests, and conditions on sequence unpredictability if the application is in security. In the present work, a literature review on cryptographically secure pseudo-random number generators (CSPRNGs) is carried out, they are implemented, and a critical analysis of their statistical quality and computational efficiency is performed. For this purpose, different programming languages will be used, and the sequences obtained will be checked by means of the NIST Statistical Test Suite (NIST STS). In addition, a user’s guide will be provided to allow the selection of one generator over another according to its statistical properties and computational implementation characteristics.

Software products cannot be delivered to the market without proper testing. Only with the help of Testing, accuracy and quality of the product improves. Test personnel cannot compromise on the quality of the product and cannot afford to miss any defects. As the product's functionality expands, so does the testcase suite, and executing all of them takes more time and work. In this discussion, we'll look at how to use a machine learning approach called Hierarchical Divisive Clustering to optimise the test suite. With this approach, all the testcases are being considered as a single cluster in the beginning and during every iteration they are separated based on the similarity. This would help execute unique testcases without compromising on the quality which would help during any regression or sanity testing. [ABSTRACT FROM AUTHOR]

Building Performance Simulation (BPS), a useful tool to assess the operational performance of buildings and systems, can often be computationally expensive. The use of BPS is cumbersome for problems where the speed of response is important, e.g. real-time control, uncertainty quantification, parametric exploration, or stock modelling. Emulators, such as those based on regression, offer a faster substitute, but their reliability can be questionable. This paper proposes seven tests to check if an emulator is a suitable replacement for simulation in practice. The tests are categorized using four criteria: accuracy, speed, generalisability, and ease of use. The tests can be included in the process of setting up an emulator-based workflow. A use case is provided for emulators based on linear and non-linear regression (Gaussian Process models). This work aims to enable a practitioner to reliably conduct performance assessment for buildings using emulators. [ABSTRACT FROM AUTHOR]

State-based testing is a challenging area in the software testing field. This paper proposes a set of algorithms to generate test cases from a state chart diagram based on various coverage criteria. The objective is to find various types of state-based faults by covering states and transitions of an object. First, the state chart diagram is transformed into an intermediate graph, State Chart Intermediate Graph (SCIG). Then, for a given coverage criteria, traversing method is applied on SCIG to generate test cases. Different trees from SCIG based on various coverage criteria are extracted for test case generation. Various coverage criteria such as All Transition (AT), Round Trip Path (RTP) and All Transition Pair (ATP) are considered. We introduce algorithms for two most efficient state-based criteria, RTP and ATP. Two case studies, Stack Operation and Vending Machine Automation system, are discussed throughout the paper. We experimentally observed that (i) AT consumes the most test resources (ii) ATP can't achieve 100% transition coverage (iii) test cases generated based on RTP is efficient, and it overcomes the transition explosion problem of AT. This analysis is beneficial in the area of semi-automatic test case generation in model-based testing. [ABSTRACT FROM AUTHOR]

Visualizing the search behavior of a series of points or populations in their native domain is critical in understanding biases and attractors in an optimization process. Distance-based many-objective optimization test problems have been developed to facilitate visualization of search behavior in a 2-D design space with arbitrarily many objective functions. Previous works have proposed a few commonly seen problem characteristics into this problem framework, such as the definition of disconnected Pareto sets and dominance resistant regions of the design space. The authors’ previous work has advanced this research further by providing a problem generator to automatically create user-defined problem instances featuring any combination of these problem features as well as newly introduced ones, such as landscape discontinuities, varying objective ranges, and neutrality. This work makes a number of additional contributions including the proposal of an enhanced, open-source feature-rich problem generator that can create user-defined problem instances exhibiting a range of problem features—some of which are newly introduced here or form extensions of existing features. A comprehensive validation of the problem generator is also provided using popular multiobjective optimization algorithms, and some problem generator settings to create instances exhibiting different challenges for an optimizer are identified. [ABSTRACT FROM AUTHOR]

Performance evaluation of optimization-based antenna design methods is essential. It is desirable to develop an efficient test suite with characteristics of antenna structures to replace the high computational cost electromagnetics (EM)-based evaluation. In this article, four typical antennas with respect to four typical objectives based on EM $S$ -parameter simulations are systematically investigated. Based on the investigation results, we propose a performance evaluation test suite allowing efficient and versatile benchmarking for antenna $S$ -parameter optimization algorithms. The test suite consists of eight analytical functions (four unimodal functions, two multimodal functions, and two compositional functions). Each function matches a typical landscape of the EM-simulated antenna problems in the 2-D case. Seven of the functions are scalable to high-dimension problems. The performances of six widely used global optimization algorithms are evaluated using the test suite. The performance evaluation results are compared to those obtained using the two EM-based antenna problems (a single antenna and a multiantenna). The results using the proposed test suite for all six optimization algorithms are consistent with those of the real EM antenna problem. The time cost for performance evaluation using the test suite is only a fraction of the cost using the real EM antenna problems. [ABSTRACT FROM AUTHOR]

For solving constrained multiobjective optimization problems (CMOPs), many algorithms have been proposed in the evolutionary computation research community for the past two decades. Generally, the effectiveness of an algorithm for CMOPs is evaluated by artificial test problems. However, after a brief review of current artificial test problems, we have found that they are not well-designed and fail to reflect the characteristics of real-world applications (e.g., small feasibility ratio). Thus, in this paper, we first propose a new constraint construction method to facilitate the systematic design of test problems. Then, on the basis of this method, we design a new test suite consisting of 14 instances, which covers diverse characteristics extracted from real-world CMOPs and can be divided into four types. Considering that the comprehensive performance comparisons among the constraint-handling techniques (CHTs) remain scarce, we choose several representative CHTs and compare their performance on our test suite. The performance comparisons identify the strengths and weaknesses of different CHTs on different types of CMOPs and provide guidelines on how to select/design a CHT in a specific scenario. [ABSTRACT FROM AUTHOR]