Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

In this Chapter, we imagine a decision maker (or a group of experts) trying to establish or examine fair procedures to combine opinions about alternatives related to different points of view.

Multiple criteria decision making

Differential Evolution (DE) is arguably one of the most powerful and versatile evolutionary optimizers for the continuous parameter spaces in recent times. Almost 5 years have passed since the first comprehensive survey article was published on DE by Das and Suganthan in 2011. Several developments have been reported on various aspects of the algorithm in these 5 years and the research on and with DE have now reached an impressive state. Considering the huge progress of research with DE and its applications in diverse domains of science and technology, we find that it is a high time to provide a critical review of the latest literatures published and also to point out some important future avenues of research. The purpose of this paper is to summarize and organize the information on these current developments on DE. Beginning with a comprehensive foundation of the basic DE family of algorithms, we proceed through the recent proposals on parameter adaptation of DE, DE-based single-objective global optimizers, DE adopted for various optimization scenarios including constrained, large-scale, multi-objective, multi-modal and dynamic optimization, hybridization of DE with other optimizers, and also the multi-faceted literature on applications of DE. The paper also presents a dozen of interesting open problems and future research issues on DE.

Recent advances in differential evolution – An updated survey

In evolutionary multiobjective optimization, maintaining a good balance between convergence and diversity is particularly crucial to the performance of the evolutionary algorithms (EAs). In addition, it becomes increasingly important to incorporate user preferences because it will be less likely to achieve a representative subset of the Pareto-optimal solutions using a limited population size as the number of objectives increases. This paper proposes a reference vector-guided EA for many-objective optimization. The reference vectors can be used not only to decompose the original multiobjective optimization problem into a number of single-objective subproblems, but also to elucidate user preferences to target a preferred subset of the whole Pareto front (PF). In the proposed algorithm, a scalarization approach, termed angle-penalized distance, is adopted to balance convergence and diversity of the solutions in the high-dimensional objective space. An adaptation strategy is proposed to dynamically adjust the distribution of the reference vectors according to the scales of the objective functions. Our experimental results on a variety of benchmark test problems show that the proposed algorithm is highly competitive in comparison with five state-of-the-art EAs for many-objective optimization. In addition, we show that reference vectors are effective and cost-efficient for preference articulation, which is particularly desirable for many-objective optimization. Furthermore, a reference vector regeneration strategy is proposed for handling irregular PFs. Finally, the proposed algorithm is extended for solving constrained many-objective optimization problems.

A Reference Vector Guided Evolutionary Algorithm for Many-Objective Optimization

Over the last three decades, a large number of evolutionary algorithms have been developed for solving multi-objective optimization problems. However, there lacks an upto-date and comprehensive software platform for researchers to properly benchmark existing algorithms and for practitioners to apply selected algorithms to solve their real-world problems. The demand of such a common tool becomes even more urgent, when the source code of many proposed algorithms has not been made publicly available. To address these issues, we have developed a MATLAB platform for evolutionary multi-objective optimization in this paper, called PlatEMO, which includes more than 50 multiobjective evolutionary algorithms and more than 100 multi-objective test problems, along with several widely used performance indicators. With a user-friendly graphical user interface, PlatEMO enables users to easily compare several evolutionary algorithms at one time and collect statistical results in Excel or LaTeX files. More importantly, PlatEMO is completely open source, such that users are able to develop new algorithms on the basis of it. This paper introduces the main features of PlatEMO and illustrates how to use it for performing comparative experiments, embedding new algorithms, creating new test problems, and developing performance indicators. Source code of PlatEMO is now available at: http://bimk.ahu.edu.cn/index.php?s=/Index/Software/index.html.

PlatEMO: A MATLAB Platform for Evolutionary Multi-Objective Optimization [Educational Forum]

This paper discusses the application of fuzzy reasoning spiking neural P systems with trapezoidal fuzzy numbers (tFRSN P systems) to fault diagnosis of power systems, where a matrix-based fuzzy reasoning algorithm based on the dynamic firing mechanism of neurons is used to develop the inference ability of tFRSN P systems from classical reasoning to fuzzy reasoning. Some case studies show the effectiveness of the presented method. We also briefly draw comparisons between the presented method and several main fault diagnosis approaches from the perspectives of knowledge representation and inference process.

/pdf/application-of-fuzzy-reasoning-spiking-neural-p-systems-to-wyzp5q8n6s.pdf

Application of Fuzzy Reasoning Spiking Neural P Systems to Fault Diagnosis

This paper proposes a novel method called multi-criterion satisfactory optimization method (MCSOM) to design infinite impulse response (IIR) filters. The core of the method is that multiple criteria can be adjusted conveniently, according to different requirements of different designers for performances of pass bands, stop bands and transition bands of IIR digital filters. The method can be employed directly in frequency domain. By designing performance specification satisfaction rate functions and synthesis satisfactory rate function of IIR digital filters, satisfactory optimization model is constructed. The model is characterized by considering satisfactory design of performance specifications and optimization of parameters of IIR digital filters, simultaneously. Furthermore, a new quantum genetic algorithm (NQGA) is also proposed to search automatically satisfactory solution in IIR digital filter design in this paper. Rapid convergence and good global search capability characterized the performance of NQGA. The effectiveness, applicability and flexibility of MCSOM based on NQGA are demonstrated by experimental results on the lowpass and bandpass IIR digital filters design. The results show that the introduced method is superior to Yule-Walker method and the method in Li Jianhua and Yin Fuliang (1996).

Multi-criterion satisfactory optimization method for designing IIR digital filters

Feature selection is always an important and difficult issue in pattern recognition, machine learning and data mining. In this paper, a novel approach called resemblance coefficient feature selection (RCFS) is proposed. Definition, properties of resemblance coefficient (RC) and the evaluation criterion of the optimal feature subset are given firstly. Feature selection algorithm using RC criterion and a quantum genetic algorithm is described in detail. RCFS can decide automatically the minimal dimension of good feature vector and can select the optimal feature subset reliably and effectively. Then the efficient classifiers are designed using neural network. Finally, to bring into comparison, 3 methods, including RCFS, sequential forward selection using distance criterion (SFSDC) and a new method of feature selection (NMFS) presented by Tiejun Lu are used respectively to select the optimal feature subset from original feature set (OFS) composed of 16 features of radar emitter signals. The feature subsets, obtained from RCFS, SFSDC and NMFS, and OFS are employed respectively to recognize 10 typical radar emitter signals in a wide range of signal-to-noise rate. Experiment results show that RCFS not only lowers the dimension of feature vector greatly and simplifies the classifier design, but also achieves higher accurate recognition rate than SFSDC, NMFS and OFS, respectively.

/pdf/resemblance-coefficient-and-a-quantum-genetic-algorithm-for-5742u12m29.pdf

Resemblance Coefficient and a Quantum Genetic Algorithm for Feature Selection

An accurate and dynamically robust state estimator is indispensable for the efficient and reliable operation of a power system. In this paper, a state estimation method based on Phasor Measurement Units (PMUs) is proposed for the real-time monitoring of power systems under various operating conditions. This PMU-based Robust Dynamic State Estimator (PRDSE) makes feasible the combination of historical measurements, obtained from the Supervisory Control And Data Acquisition (SCADA) system, with present more precision measurements obtained from PMUs. A new state accuracy-based weighting function is proposed to increase the robustness when the system encounters a large unwanted disturbance. Several IEEE test systems under normal and dynamic operation conditions are used to demonstrate the high performance of the PRDSE. Numerical results show the effectiveness and robustness of the PRDSE.

PMU based Robust Dynamic State Estimation method for power systems

To decrease the computational complexity and improve the search capability of quantum-inspired evolutionary algorithm based on P systems (QEPS), a real-observation QEPS (RQEPS) was proposed RQEPS is a hybrid algorithm combining the framework and evolution rules of P systems with active membranes and real-observation quantum-inspired evolutionary algorithm (QEA) The RQEPS involves a dynamic structure including membrane fusion and division The membrane fusion is helpful to enhance the information communication among individuals and the membrane division is beneficial to reduce the computational complexity An NP-complete problem, the time-frequency atom decomposition of noised radar emitter signals, is employed to test the effectiveness and practical capabilities of the RQEPS The experimental results show that RQEPS is superior to QEPS, the greedy algorithm and binary-observation QEA in terms of search capability and computational complexity

/pdf/an-improved-membrane-algorithm-for-solving-time-frequency-18ajs506ys.pdf

Gexiang Zhang

Papers

Application of Fuzzy Reasoning Spiking Neural P Systems to Fault Diagnosis

Multi-criterion satisfactory optimization method for designing IIR digital filters

Resemblance Coefficient and a Quantum Genetic Algorithm for Feature Selection

PMU based Robust Dynamic State Estimation method for power systems

An improved membrane algorithm for solving time-frequency atom decomposition