Showing papers on "Artificial immune system published in 2001"
TL;DR: It is argued that this new resource-based mechanism is a large step forward in making AISs a viable contender for effective unsupervised machine learning and allows for not just a one shot learning mechanism, but a continual learning model to be developed.
Abstract: This paper presents a resource limited artificial immune system (RLAIS) for data analysis. The work presented here builds upon previous work on artificial immune systems (AIS) for data analysis. A population control mechanism, inspired by the natural immune system, has been introduced to control population growth and allow termination of the learning algorithm. The new algorithm is presented, along with the immunological metaphors used as inspiration. Results are presented for Fisher Iris data set, where very successful results are obtained in identifying clusters within the data set. It is argued that this new resource-based mechanism is a large step forward in making AISs a viable contender for effective unsupervised machine learning and allows for not just a one shot learning mechanism, but a continual learning model to be developed.
349 citations
Proceedings Article•
07 Jul 2001
TL;DR: It is suggested that the most appropriate use of negative selection in the AIS is as a filter for invalid detectors, not the generation of competent detectors.
Abstract: This paper investigates the role of negative selection in an artificial immune system (AIS) for network intrusion detection. The work focuses on the use of negative selection as a network traffic anomaly detector. The results of the negative selection algorithm experiments show a severe scaling problem for handling real network traffic data. The paper concludes by suggesting that the most appropriate use of negative selection in the AIS is as a filter for invalid detectors, not the generation of competent detectors.
221 citations
27 May 2001
TL;DR: The embedded negative selection operator plays an important role in the AIS by helping it to maintain a low false positive detection rate, and how to choose appropriate detector and antigen sample sizes is investigated.
Abstract: The paper describes research towards the use of an artificial immune system (AIS) for network intrusion detection. Specifically, we focus on one significant component of a complete AIS, static clonal selection with a negative selection operator, describing this system in detail. Three different data sets from the UCI repository for machine learning are used in the experiments. Two important factors, the detector sample size and the antigen sample size, are investigated in order to generate an appropriate mixture of general and specific detectors for learning non-self antigen patterns. The results of series of experiments suggest how to choose appropriate detector and antigen sample sizes. These ideal sizes allow the AIS to achieve a good non-self antigen detection rate with a very low rate of self antigen detection. We conclude that the embedded negative selection operator plays an important role in the AIS by helping it to maintain a low false positive detection rate.
207 citations
29 Nov 2001
TL;DR: An investigation has been undertaken to repeat previous work on an artificial immune system for data analysis called AINE (Artificial Immune Network) and it is argued that AINE is an effective data-mining algorithm.
Abstract: An investigation has been undertaken to repeat previous work on an artificial immune system for data analysis called AINE (Artificial Immune Network).The previous work was limited to testing the algorithm on relatively small data sets. The aim of this investigation is two fold,firstly to corroborate the results presented in previous work and secondly, to test the algorithm on a larger and more complex data set. A new re-implementation of AINE is then described and differences in behaviour are identified and explained. It is argued that the behaviourseen in the new implementation is more accurate than that seen in previous work and an in-depth analysis of the algorithm structure is undertaken in order to confirm theseobservations. The algorithm is also tested on new data and the results of this are presented. Comparisons are draw with other similar techniques for data mining and it is argued that AINE is an effective data-mining algorithm.
75 citations
01 Sep 2001
TL;DR: This chapter describes the physiology of the immune system and provides a general introduction to Artificial Immune Systems, and concludes with an evaluation of the current and future contributions of Artificial Immunes Systems in Data Mining.
Abstract: The immune system is highly distributed, highly adaptive, self-organising in nature, maintains a memory of past encounters, and has the ability to continually learn about new encounters. From a computational viewpoint, the immune system has much to offer by way of inspiration. Recently there has been growing interest in the use of the natural immune system as inspiration for the creation of novel approaches to computational problems; this field of research is referred as Immunological Computation (IC) or Artificial Immune Systems (AIS). This chapter describes the physiology of the immune system and provides a general introduction to Artificial Immune Systems. Significant applications that are relevant to data mining, in particular in the areas of machine learning and data analysis are discussed in detail. Attention is paid both to the salient characteristics of the application and the details of the algorithms. This chapter concludes with an evaluation of the current and future contributions of Artificial Immune Systems in Data Mining.
66 citations
12 Jul 2001
TL;DR: It is shown a finite state machine can be provided with a hardware immune system to provide a novel form of fault detection giving the ability to detect every faulty state during a normal operating cycle, called immunotronics.
Abstract: Since the advent of fault tolerance in the 1960s, numerous techniques have been developed to increase the reliability of safety critical and space borne missions. In the last decade novel approaches to this field have sought inspiration from nature in the form of evolutionary and developmental forms of fault tolerance. In nature an additional inspiration axis exists in the form of learning. The body's own immune system uses a form of learning to maintain reliable operation in the body even in the presence of invaders. This has only recently been applied as a computational technique in the form of artificial immune systems (AIS). This paper demonstrates a new application of AIS with an immunologically inspired approach to fault tolerance. It is shown a finite state machine can be provided with a hardware immune system to provide a novel form of fault detection giving the ability to detect every faulty state during a normal operating cycle. We call this immunotronics.
51 citations
TL;DR: An intelligent agent will be described for task allocation in a heterogeneous computing environment and some of its functionalities will be exploited in designing agent-based parallel and distributed control systems.
41 citations
27 May 2001
TL;DR: Four of these new branches of research are outlined: creative evolutionary systems, computational embryology, evolvable hardware and artificial immune systems, showing how they aim to extend the capabilities of EC.
Abstract: In the last five years, the field of evolutionary computation (EC) has seen a resurgence of new ideas, many stemming from new biological inspirations. The paper outlines four of these new branches of research: creative evolutionary systems, computational embryology, evolvable hardware and artificial immune systems, showing how they aim to extend the capabilities of EC. Recent, unpublished results by researchers in each area at the Department of Computer Science, UCL are provided.
41 citations
25 Jul 2001
TL;DR: Simulation results reveal that immune network algorithms are effective for searching for optimal control against disturbances and an possibility of applying the flexible arbitration abilities of an artificial immune network has been suggested for the PID controller tuning.
Abstract: Suggests that immune algorithms can be used effectively for the tuning of PID control structures for nonlinear processes. The controller's attribute behavior mechanism in the plant and the artificial immune system have certain similarities, since both systems deal with various attribute inputs and outputs through interactions among multiple-attribute modules. Since antibodies communicate with each other among different species of antibodies/B-cells through the stimulation and suppression chains among the antibodies that form a large-scale network, the artificial immune network system always has a new parallel decentralized processing mechanism for various situations. In addition to that, the structure of the network is not fixed but varies continuously, i.e. the artificial immune network flexibly self-organizes according to dynamic changes of the external environment. On the other hand, a number of tuning methods on the PID controller have been considered but, with only the P, I and D parameters, it is very difficult to control a plant with complex dynamics, such as large dead time, inverse response and highly nonlinear characteristics. An possibility of applying the flexible arbitration abilities of an artificial immune network has been suggested for the PID controller tuning. Simulation results reveal that immune network algorithms are effective for searching for optimal control against disturbances.
38 citations
TL;DR: The model is shown to discover and maintain coverage of the diverse patterns through mechanism of evolution and mutation and to study pattern recognition processes and learning.
34 citations
01 Jan 2001
TL;DR: A biologically inspired method based on artificial immune systems for supervised learning for multilayer feedforward neural networks is proposed, and its performance is compared to that produced by other approaches already suggested in the literature.
Abstract: The initial weight vector to be used in supervised learning for multilayer feedforward neural networks has a strong influence in the learning speed and in the quality of the solution obtained after convergence. An inadequate initial choice may cause the training process to get stuck in a poor local minimum, or to face abnormal numerical problems. In this paper, we propose a biologically inspired method based on artificial immune systems. This new strategy is applied to several benchmark and real-world problems, and its performance is compared to that produced by other approaches already suggested in the literature.
TL;DR: It is argued that biological immune systems share a number of similarities with ecological economic systems in terms of function, which include the system's ability to recognize harmful invasions, design measures to control and destroy theseInvasions, and remember successful response strategies.
Abstract: A new perspective for studying the complex interactions between human activities and ecosystems is proposed. It is argued that biological immune systems share a number of similarities with ecological economic systems in terms of function. These similarities include the system's ability to recognize harmful invasions, design measures to control and destroy these invasions, and remember successful response strategies. Studying both the similarities and the differences between immune systems and ecological economic systems can provide new
07 Oct 2001
TL;DR: A computational implementation of negative selection of immune system along with genetic algorithm to perform a color image classification task and the percentage of correct classification is obtained.
Abstract: An artificial immune system is applied to a pattern recognition problem. Artificial immune systems (AIS) possess nonlinear classification properties along with the biological properties such as self/non-self identification, positive and negative selection. In this paper, we propose a computational implementation of negative selection of immune system along with genetic algorithm to perform a color image classification task. The images used for classification are finished wooden components. The wooden components are kitchen cabinets manufactured by American Woodmark Corporation. The images are obtained from Virginia Polytechnic Institute and State University. The classification addresses the average RGB values of the images and the percentage of correct classification is obtained for a set of test and sample images.
Book Chapter•
01 Apr 2001
TL;DR: The usefulness of using AIN for exploratory visualisation is investigated and an explanation of how aiVIS operates is presented.
Abstract: The field of Artificial Immune Systems (AIS) is the use of the natural immune system as a metaphor for solving computational problems. A novel unsupervised machine-learning algorithm, inspired by the immune system, has been developed called AINE. Using various immunological metaphors, AINE evolves a network of objects, known as an Artificial Immune Network (AIN) that is a diverse representation of the data set being learnt. The results of AINE are visualised in a specially developed tool (aiVIS), which allows the user to interact with the network to perform exploratory analysis. aiVIS presents AINs in such as way as to build up an understanding of the make up of the data set, learning about subtle patterns and clusters within the data set and links between clusters. Unclassified items can then be introduced into the network so to further enhance the exploratory nature of the AIN. This paper provides an overview of the learning algorithm, but concentrates on the visualisation aspect of the work. The usefulness of using AIN for exploratory visualisation is investigated and an explanation of how aiVIS operates is presented.
21 May 2001
TL;DR: The exploration of the basic principles that govern an immune system and the potential implementation of these principles in a multi-agent ISS of a heterogeneous computer network is explored.
Abstract: Advanced information security systems (ISS) play an ever-increasing role in the information assurance in global computer networks. Dependability of ISS is being achieved by the enormous amount of data processing that adversely affects the overall network performance. Modern ISS architecture is viewed as a multi-agent system comprising a number of semi-autonomous software agents designated to prevent particular kinds of threats and suppress specific types of attacks without burdening the network. The high efficiency of such a system is achieved by establishing the principles of successful individual and cooperative operation of particular agents. Such principles, evolved during evolution, are known to be implemented in biological immune systems. The aim of this paper is the exploration of the basic principles that govern an immune system and the potential implementation of these principles in a multi-agent ISS of a heterogeneous computer network.
01 Jan 2001
TL;DR: In this article, the authors present an assessment of the resource limited artificial immune system known as AINE, which is a continuation of previous work to develop an artificial immune systems for data analysis.
Abstract: This report presents an assessment of the resource limited artificial immune system known as AINE. This work is a continuation of previous work to develop an artificial immune system for data analysis. A brief introduction to the fundamentals of immunology is given followed by a review of the previous work on AINE. Discrepancies in the original work are identified and revisions are made. Results are then presented for a simulated data set and for the Fisher Iris data set. Comparisons are then drawn between the revised algorithm and the original version of AINE. Results of parameter adjustment on the revised version confirm those results in the previous work. Additionally, trends in the evolution of the networks are identified and analysed. These tests show new behaviour in the revised AINE and from which it can be concluded that AINE no longer shows the possibility of continual learning, but exhibits characteristics of optimisation, which are identified for future study.
01 May 2001
TL;DR: It can be concluded that AINE no longer shows the possibility of continual learning, but exhibits characteristics of optimisation, which are identified for future study.
Abstract: This report presents an assessment of the resource limited artificial immune system known as AINE. This work is a continuation of previous work to develop an artificial immune system for data analysis. A brief introduction to the fundamentals of immunology is given followed by a review of the previous work on AINE. Discrepancies in the original work are identified and revisions are made. Results are then presented for a simulated data set and for the Fisher Iris data set. Comparisons are then drawn between the revised algorithm and the original version of AINE. Results of parameter adjustment on the revised version confirm those results in the previous work. Additionally, trends in the evolution of the networks are identified and analysed. These tests show new behaviour in the revised AINE and from which it can be concluded that AINE no longer shows the possibility of continual learning, but exhibits characteristics of optimisation, which are identified for future study.
21 May 2001
TL;DR: A biological approach to information security based on a rigorous mathematical notion of formal immune network, which could be considered as an alternative to the wide spread artificial neural networks or intelligent agents.
Abstract: We propose a biological approach to information security based on a rigorous mathematical notion of formal immune network. According to our previous developments, such networks possess all the main capabilities of artificial intelligence system, and could be considered as an alternative to the wide spread artificial neural networks or intelligent agents. We consider also the main distinctions of our approach from the modern information security by agent-based modeling and artificial immune systems.
18 Jun 2001
TL;DR: A simple and easy to implement algorithm for multimodal function optimization based on clonal selection and programmed cell death mechanisms taken from natural immune system is proposed.
Abstract: A simple and easy to implement algorithm for multimodal function optimization is proposed. It is based on clonal selection and programmed cell death mechanisms taken from natural immune system. Empirical results confirming its usability are presented, and review of other related approaches is given.
01 Sep 2001
TL;DR: In this paper, a simplified incremental approach integrated with the maximum entropy principle and an instantaneous feedback mechanism is proposed to reorganize the system's parameters simultaneously, which can achieve robustness and efficiency in identifying complex nonlinear systems.
Abstract: In this paper, a novel non-linear system identification methodology is developed employing the features of the artificial immune system. A simplified incremental approach integrated with the maximum entropy principle and an instantaneous feedback mechanism is proposed to reorganize the system's parameters simultaneously. To verify and demonstrate the effectiveness of the proposed algorithm, a simulation example on a two-link robot was studied. This algorithm can achieve robustness and efficiency in identifying complex non-linear systems. The simulation results show that the identified immune models are robust to noise and various uncertainties in the robot dynamics.
18 Apr 2001
TL;DR: A prototype of a new model for performing clustering in large, non-static databases is presented using a coevolutionary genetic algorithm that runs continuously in order to dynamically track clusters in the data.
Abstract: In this paper we present a prototype of a new model for performing clustering in large, non-static databases. Although many machine learning algorithms for data clustering have been proposed, none appear to specifically address the task of clustering moving data. The model we describe combines features of two existing computational models -- that of Artificial Immune Systems (AIS) and Sparse Distributed Memories (SDM). The model is evolved using a coevolutionary genetic algorithm that runs continuously in order to dynamically track clusters in the data. Although the system is very much in its infancy, the experiments conducted so far show that the system is capable of tracking moving clusters in artificial data sets, and also incorporates some memory of past clusters. The results suggest many possible directions for future research.
Dissertation•
30 Nov 2001
TL;DR: Two solutions to a multi-agent herding problem are presented and a comparative study of the Q-learning solution and the immune network solution is done on important aspects such as computation requirements, predictability, and convergence.
Abstract: " Multi-Agent systems " is a topic for a lot of research, especially research involving strategy, evolution and cooperation among various agents. Various learning algorithm schemes have been proposed such as reinforcement learning and evolutionary computing. In this thesis two solutions to a multi-agent herding problem are presented. One solution is based on Q-learning algorithm, while the other is based on modeling of artificial immune system. Q-learning solution for the herding problem is developed, using region-based local learning for each individual agent. Individual and batch processing reinforcement algorithms are implemented for non-cooperative agents. Agents in this formulation do not share any information or knowledge. Issues such as computational requirements, and convergence are discussed. An idiotopic artificial immune network is proposed that includes individual B-cell model for agents and T-cell model for controlling the interaction among these agents. Two network models are proposed – one for evolving group behavior/strategy arbitration and the other for individual action selection. i A comparative study of the Q-learning solution and the immune network solution is done on important aspects such as computation requirements, predictability, and convergence. ii ACKNOWLEDGMENTS This is a major milestone in my life, for which I would like to thank my parents. Their love and constant support over the years have been the most important factors in my achievements and my life. My deepest appreciation goes to my advisor, Dr. Pushkin Kachroo, for his valuable assistance, guidance, and encouragement in bringing this research work to a successful completion. He showed extraordinary patience and was always available for friendly advice. I am also grateful to my thesis committee members, Dr. Hugh VanLandingham and Dr. Will Saunders, for their advise and help. I would like to thank all my colleagues and friends that I have acquired throughout my years at Virginia Tech. It is their company and all great moments that I shared with them-that enabled me to successfully complete my research work. There are many others who were indirectly supportive to my work and I would like to thank all of them from my heart.
01 Jan 2001
TL;DR: The relative biological background and main biological tools of DNA computing are introduced, and some DNA computing models and algorithms are provided to explore the ideas and methods for integrating DNA computing with evolutionary computation, fuzzy control, neural networks and artificial immune systems in soft computing technologies.
Abstract: Recently, DNA computing arouses wide interests of many researchers, particularly, the integration of DNA computing and soft computing. In this paper, we first introduce the relative biological background and main biological tools of DNA computing. Then, we provide some DNA computing models and algorithms. After that, we explore the ideas and methods for integrating DNA computing with evolutionary computation, fuzzy control, neural networks and artificial immune systems in soft computing technologies.
Journal Article•
TL;DR: This paper proposes to provide autonomous adaptability in communication endsystem with OpenWebServer/iNexus, which is both a web server and an object-oriented framework to tailer various web services and applications, and its design is inspired by the natural immune system.
10 Sep 2001
TL;DR: The immune system and its adaptive properties are described and a simple artificial immune system (AIS) model based on the clonal selection theory is presented and fruitfully applied to the problem of fuzzy resource identification.
Abstract: In this report the immune system and its adaptive properties are described and a simple artificial immune system (AIS) model based on the clonal selection theory is presented. This immune system model is demonstrated to be capable of learning the structure of novel antigens, of memory for previously encountered antigens, and of being able to use its memory to respond more efficiently to antigens related to ones it has previously seen (cross-reactivity). The learning, memory and cross-reactivity of the AIS are fruitfully applied to the problem of fuzzy resource identification. Interesting antigen/antibody relationships are also identified.
Proceedings Article•
07 Jul 2001TL;DR: Good statistical results using historical metrics as well as a new spatial/temporal metric are obtained, thereby making the modified memetic algorithm a viable option for solving the PSP problem.
Abstract: Considerable research has been presented to develop a generalized technique to predict a polypeptide's molecular structure given its amino acid sequence. This is also known as the Protein Structure Prediction (PSP) problem which has direct applications to many scientific, medical, and engineering disciplines. Previous research with Evolutionary Algorithms (EAs) to minimize the empirical CHARMM protein energy model and generation of the associated protein structure is extended using the fast messy genetic algorithm improved through the use of secondary protein structure information integrated with artificial immune system concepts. Good statistical results using historical metrics as well as a new spatial/temporal metric are obtained, thereby making the modified memetic algorithm a viable option for solving the PSP problem.
Journal Article•
TL;DR: In this article, a method of cooperative control and selection of group behavior strategy based on immune system in distributed autonomous robotic systems (DARS) is proposed, which can be applied to decision making of optimal swarm behavior in dynamically changing environment.
Abstract: In this paper, we propose a method of cooperative control (T-cell modeling) and selection of group behavior strategy (B-cell modeling) based on immune system in distributed autonomous robotic systems (DARS). Immune system is living body's self-protection and self-maintenance system. These features can be applied to decision making of optimal swarm behavior in dynamically changing environment For applying immune system to DARS, a robot is regarded as a B-cell, each environmental condition as an antigen, a behavior strategy as an antibody and control parameter as a T-cell respectively. The executing process of proposed method is as follows. When the environmental condition changes, a robot selects an appropriate behavior strategy. And its behavior strategy is stimulated and suppressed by other robot using communication. Finally much stimulated strategy is adopted as a swarm behavior strategy. This control scheme is based on clonal selection and idiotopic network hypothesis. And it is used for decision making of optimal swarm strategy. By T-cell modeling, adaptation ability of robot is enhanced in dynamic environments.
23 Oct 2001
TL;DR: A brief review of one of nature's most elegant, complex adaptive systems, Immunity to pathogenic organisms.
Abstract: Immunity to pathogenic organisms is a complex process involving interacting factors within the immune system including circulating cells, tissues and soluble chemical mediators. Both the efficiency and adaptive responses of the immune system in a dynamic, often hostile, environment are essential for maintaining our health and homeostasis. This paper will present a brief review of one of nature's most elegant, complex adaptive systems.
Journal Article•
TL;DR: In this paper, a method of cooperative control and selection of group behavior strategy based on immune system in distributed autonomous robotic system (DARS) is proposed, where a robot is regarded as a B-cell, each environmental condition as an antigen, behavior strategy as an antibody and control parameter as a T-cell respectively.
Abstract: In this paper, we propose a method of cooperative control (T-cell modeling) and selection of group behavior strategy (B-cell modeling) based on immune system in distributed autonomous robotic system (DARS). Immune system is living body's self-protection and self-maintenance system. These features can be applied to decision making of optimal swarm behavior in dynamically changing environment. For applying immune system to DARS, a robot is regarded as a B-cell, each environmental condition as an antigen, a behavior strategy as an antibody and control parameter as a T-cell respectively. The executing process of proposed method is as follows. When the environmental condition changes, a robot selects an appropriate behavior strategy. And its behavior strategy is stimulated and suppressed by other robot using communication. Finally much stimulated strategy is adopted as a swarm behavior strategy. This control school is based on clonal selection and idiotopic network hypothesis. And it is used for decision making of optimal swarm strategy. By T-cell modeling, adaptation ability of robot is enhanced in dynamic environments.