Genetics-based Machine Learning Research Articles

Hybridization of Fuzzy GBML Approaches for Pattern Classification Problems

We propose a hybrid algorithm of two fuzzy genetics-based machine learning approaches (i.e., Michigan and Pittsburgh) for designing fuzzy rule-based classification systems. First, we examine the search ability of each approach to efficiently find fuzzy rule-based systems with high classification accuracy. It is clearly demonstrated that each approach has its own advantages and disadvantages. Next, we combine these two approaches into a single hybrid algorithm. Our hybrid algorithm is based on the Pittsburgh approach where a set of fuzzy rules is handled as an individual. Genetic operations for generating new fuzzy rules in the Michigan approach are utilized as a kind of heuristic mutation for partially modifying each rule set. Then, we compare our hybrid algorithm with the Michigan and Pittsburgh approaches. Experimental results show that our hybrid algorithm has higher search ability. The necessity of a heuristic specification method of antecedent fuzzy sets is also demonstrated by computational experiments on high-dimensional problems. Finally, we examine the generalization ability of fuzzy rule-based classification systems designed by our hybrid algorithm.

Research on the Construction of Fuzzy Classifier System for Multidimensional Pattern Classification Using Genetic Algorithms

This paper discusses the application and performance of multidimensional pattern classification problems using Michigan approach based on fuzzy genetics-based machine learning mechanism, and proposes a new approach. In the approach, each fuzzy if-then rule is handled as an individual, and a fitness value is assigned to it. The approach not only retrieves fuzzy if-then rules, but also tunes the membership functions of each dimension, meanwhile the selection mechanism based on the similarity of individuals is involved to reduce the high selective pressure, keep the diversity of population, and avoid the premature convergence problem consequently. Finally the experiments prove that the approach has a better correct classification rate and a better adaptability on multidimensional pattern classification problems.

Comparing learning classifier systems and Genetic Programming: a case study

Genetic Algorithms has given rise to two new fields of research where (global) optimisation is of crucial importance: ‘ genetic based machine learning’ (GBML) and ‘ genetic programming’ (GP). An advanced implementation of GBML (Fuzzy Efficiency based Classifier System, FECS, developed by the authors) and GP (as defined by Koza) are both applied to the case study ‘fibre-to-yarn production process’. Results for both systems are presented and compared. Finally, the GP generated equations are transformed into rule-sets similar to those obtained from FECS.

A Genetics-Based Machine Learning Approach for Realtime Scheduling

In this paper, we adopt a genetic-based machine learning (GBML) approach to a realtime scheduling problem in which several products are to be assigned to one of the buffers, and propose a method of generating and selecting rules for assigning each product to a desirable buffer. In applying the GBML, we use the Pitt approach, where the set of rules (rule-set) is represented symbolically as an individual of genetic algorithms, and the fitness of an individual is calculated based on the total cost required for transporting, operating and keeping of whole products. Through some computational experiments, the effectiveness and the possibility of our approach is investigated.

분류자 시스템을 이용한 축구 로봇의 행동 전략

Learning Classifier System (LCS) finds a new rule set using genetic algorithm (GA). In this paper, The Zeroth Level Classifier System (ZCS) is applied to evolving the strategy of a robot soccer simulation game (SimuroSot), which is a state varying dynamical system changed over time, as GBML (Genetic Based Machine Learning) and we show the effectiveness of the proposed scheme through the simulation of robot soccer.

Genetic programming: principles and applications

Genetic algorithms (GA) has given rise to two new fields of research where (global) optimisation is of crucial importance: ‘ genetic based machine learning’ (GBML) and ‘ genetic programming’ (GP). An introduction by the authors to GA and GBML was given in two previous papers (Eng. Appl. Artif. Intell. 9(6) (1996) 681; Eng. Appl. Artif. Intell. 13(4) (2000) 381). In this paper, the last domain (GP) will be introduced, thereby making up a trilogy which gives a general overview of the whole field. In this third part, an overview will be given of the basic concepts of GP as defined by Koza. A first (educational) example of GP is given by solving a simple symbolic regression of a sinus function. Finally, a more complex application is presented in which GP is used to construct the mathematical equations for an industrial process. To this end, the case study ‘fibre-to-yarn production process’ is introduced. The goal of this study is the automatic development of mathematical equations for the prediction of spinnability and (possible) resulting yarn strength. It is shown that (relatively) simple equations can be obtained which describe accurately 90% of the fibre-to-yarn database.

Comparing Learning Classifier Systems and Genetic Programming: A Case Study

ABSTRACT Genetic Algorithms has given rise to two new fields of research where (global) optimisation is of crucial importance: ' Genetic based Machine Learning ' (GBML) and ' Genetic Programming ' (GP). An advanced implementation of GBML (Fuzzy Efficiency based Classifier System, FECS, developed by the authors) and Genetic Programming (as defmed by Koza) are both applied to the case study 'fibre-to-yarn production process'. Results for both systems are presented and compared. Finally, the GP generated equations are transformed into rule-sets similar to those obtained from FECS.

Three-objective genetics-based machine learning for linguistic rule extraction

Abstract This paper shows how a small number of linguistically interpretable fuzzy rules can be extracted from numerical data for high-dimensional pattern classification problems. One difficulty in the handling of high-dimensional problems by fuzzy rule-based systems is the exponential increase in the number of fuzzy rules with the number of input variables. Another difficulty is the deterioration in the comprehensibility of fuzzy rules when they involve many antecedent conditions. Our task is to design comprehensible fuzzy rule-based systems with high classification ability. This task is formulated as a combinatorial optimization problem with three objectives: to maximize the number of correctly classified training patterns, to minimize the number of fuzzy rules, and to minimize the total number of antecedent conditions. We show two genetic-algorithm-based approaches. One is rule selection where a small number of linguistically interpretable fuzzy rules are selected from a large number of prespecified candidate rules. The other is fuzzy genetics-based machine learning where rule sets are evolved by genetic operations. These two approaches search for non-dominated rule sets with respect to the three objectives.

An Application of Genetics-Based Machine Learning Approach to a Realtime Assignment Problem

An Application of Genetics-Based Machine Learning Approach to a Realtime Assignment Problem

F-1231 擬似生態系における群ロボットのGAに基づく行動学習(G15-2 ロボットの知能化制御)(G15 ロボティクス・メカトロニクス部門一般講演)

This paper deals with behavioral evolution of multiple robots in a quasi-ecosystem. An ecosystem model composed of insects and plants, which are in a relationship of parasitism, is simulated on the discrete cell space. In this ecosystem, the plants become easy to be eliminated as the population size of the insects increases. Consequently, it is required to maintain numerical balance of plants and insects in the quasi-ecosystem. Therefore, multiple robots are introduced to remove some insects from the quasi-ecosystem. However, if the robots eliminate all of insects, the viruses will eliminate the plants owing to diseases. In this ecosystem with complicated relationship, the robots should acquire strategies to maintain plants. In this paper, we use simple if-then rules and apply genetics-based machine learning for acquiring a strategy for removing insects. Furthermore, we show several simulation results of behavior learning of multiple robots.

An implementation of genetic algorithms for rule based machine learning

Genetic algorithms have given rise to two new fields of research where (global) optimisation is of crucial importance: ‘Genetic Programming’ and ‘Genetic based Machine Learning’ (GBML). In this paper the second domain (GBML) will be introduced. An overview of one of the first GBML implementations by Holland, also known as the Learning Classifier Systems (LCS) will be given. After describing and solving a well-known basic (educational) problem a more complex application of GBML is presented. The goal of this application is the automatic development of a rule set for an industrial production process. To this end, the case study on generating a rule set for predicting the spinnability in the fibre-to-yarn production process will be presented. A largely modified LCS, called Fuzzy Efficiency based Classifier System (FECS), originally designed by one of the authors, is used to solve this problem successfully.

Classifier systems in combat: two-sided learning of maneuvers for advanced fighter aircraft

This paper reports the continuing results of a project where a genetics-based machine learning system acquires rules for novel fighter combat maneuvers through simulation. In this project, a genetics-based machine learning system was implemented to generate high angle-of-attack air combat tactics for advanced fighter aircraft. This system, which was based on a learning classifier system approach, employed a digital simulation model of one-versus-one air combat, and a genetic algorithm, to develop effective tactics for the X-31 experimental fighter aircraft. Previous efforts with this system showed that the resulting maneuvers allowed the X-31 to successfully exploit its post-stall capabilities against a conventional fighter opponent. This demonstrated the ability of the genetic learning system to discover novel tactics in a dynamic air combat environment. The results gained favorable evaluation from fighter aircraft test pilots. However, these pilots noted that the static strategy employed by the X-31's opponent was a limitation. In response to these comments, this paper reports new results with two-sided learning, where both aircraft in a one-versus-one combat scenario use genetics-based machine learning to adapt their strategies. The experiments successfully demonstrate both aircraft developing objectively interesting strategies. However, the results also point out the complexity of evaluating results from mutually adaptive players, due to the red queen effect. These complexities, and future directions of the project, are discussed in the paper's conclusions.

An adaptive classifier system tree for extending genetics-based machine learning in a dynamic environment

An autonomous agent should possess the ability to adapt its cognition structure to a dynamically changing environment. This ability may be achieved when au- tonomous agents interact with the environment. In this paper, an adaptive classifier system tree is proposed for extending genetics-based machine learning in a dynamic environment. The architecture has the properties of self- similarity and self-organization. When environmental changes are inspected, the autonomous agent can adapt its cognition structure to the new environment so that cognition can be achieved with great efficiency. After description of the dynamic structure and the principle of the structure's self-organization, some experiments illustrating how the architecture works are described and discussed.

Fuzzy Genetics-Based Machine Learning for Pattern Classification

Fuzzy Genetics-Based Machine Learning for Pattern Classification

Performance evaluation of fuzzy classifier systems for multidimensional pattern classification problems

We examine the performance of a fuzzy genetics-based machine learning method for multidimensional pattern classification problems with continuous attributes. In our method, each fuzzy if-then rule is handled as an individual, and a fitness value is assigned to each rule. Thus, our method can be viewed as a classifier system. In this paper, we first describe fuzzy if-then rules and fuzzy reasoning for pattern classification problems. Then we explain a genetics-based machine learning method that automatically generates fuzzy if-then rules for pattern classification problems from numerical data. Because our method uses linguistic values with fixed membership functions as antecedent fuzzy sets, a linguistic interpretation of each fuzzy if-then rule is easily obtained. The fixed membership functions also lead to a simple implementation of our method as a computer program. The simplicity of implementation and the linguistic interpretation of the generated fuzzy if-then rules are the main characteristic features of our method. The performance of our method is evaluated by computer simulations on some well-known test problems. While our method involves no tuning mechanism of membership functions, it works very well in comparison with other classification methods such as nonfuzzy machine learning techniques and neural networks.

遺伝的機械学習アルゴリズム

遺伝的機械学習アルゴリズム

Constructing petri net models using genetic search

The problem considered is that of constructing a Petri Net model of a particular device, process, or system described by observations of its interactions with its environment. The algorithm proposed for achieving a solution employs the principles of genetic search. Also presented is a detailed investigation of its operation, thereby affording a theoretical justification of its design. The expressive and representational power of Petri Nets renders them ideally suited to the modelling of many complex event systems. As a modelling language, they directly support the intrinsically difficult concepts of concurrent and parallel activities, synchronisation of events and the distribution of various resources. However, the development of a model of a significant system is laborious and circumstances of limited knowledge of the system's internal structure compound the difficulty. By composing the perceived stimuli and responses of the system under study as a set of behavioural requirements, the Petri Net construction problem can be examined. The inherently difficult nature of this problem renders most other approaches inapplicable or inadequate; the quest for a satisfactory solution leads instead to the development of an algorithm employing genetic search technology. Built around a genetics-based machine learning architecture, the first algorithm developed shows a deficiency in its dependence on the particular order in which various options are explored. Alleviating this difficulty through a simple modification may also hold a lesson for other classifier systems.

A genetic-based machine learning system to discover the diagnostic rules for female urinary incontinence

A machine learning system named Galactica has been developed which uses a genetic algorithm to discover the rules for an expert system from databases. Galactica devised accurate diagnostic rules for female urinary incontinence from difficult heterogeneous data. The percentages of correctly classified stress, mixed and sensory urge incontinence testing cases were 89, 86 and 87%, respectively. However, these rules were rather general, consisting of 4–6 out of 13 conditions available in the data. Diagnostic rules for stress and mixed incontinence extracted from straightforward homogeneous data were highly accurate, classifying 100% of testing cases correctly as well as being specific, having from 10 to 11 conditions. More specific, but less accurate, rules were found from heterogeneous data with a biased fitness function. All of the rules were correct, i.e. every condition in the rules had the expected value specified by the expert. Although, Galactica achieved a slightly better classification than the discriminant analysis, it is argued that the genetic approach is better than the statistical one, due to symbolic rules being comprehensible, whereas understanding a complex mathematical model requires statistical expertise.

A new approach to genetic based machine learning for efficient local improvement and its application to a graphic problem

This paper presents a new approach to genetic based machine learning (GBML). The new approach utilizes mechanisms of genetic recombination in bacterial genetics. The new approach is efficient in improving local portions of chromosomes. A simulation of a simple graphic problem using the new approach is done and the effectiveness of the new approach is shown.

Comparison of the Michigan and Pittsburgh approaches to the design of fuzzy classification systems

Fuzzy systems based on fuzzy if-then rules have been applied to various problems. The main application area has been fuzzy control problems. In many cases, such fuzzy systems can handle only a few input variables. This is because the number of fuzzy if-then rules exponentially increases as the number of input variables increases. In this paper, we try to design fuzzy classification systems based on fuzzy if-then rules for multidimensional pattern classification problems with many attributes. For designing such fuzzy classification systems, we compare two frameworks in the area of genetics-based machine learning: the Michigan approach and the Pittsburgh approach. The performance of fuzzy rule-based classification systems is also compared with that of various pattern classification methods. In computer simulations, we use a wine classification problem with 13 attributes, a cancer diagnosis problem with 9 attributes, and a credit approval problem with 14 attributes. © 1997 Scripta Technica, Inc. Electron Comm Jpn Pt 3, 80(12): 10–19, 1997