Showing papers by "Lakhmi C. Jain published in 2005"

Evolutionary Multiobjective Optimization

Abstract: Very often real-world applications have several multiple conflicting objectives. Recently there has been a growing interest in evolutionary multiobjective optimization algorithms that combine two major disciplines: evolutionary computation and the theoretical frameworks of multicriteria decision making. In this introductory chapter, some fundamental concepts of multiobjective optimization are introduced, emphasizing the motivation and advantages of using evolutionary algorithms. We then lay out the important contributions of the remaining chapters of this volume.

Evolutionary multiobjective optimization : theoretical advances and applications

Abstract: Evolutionary Multiobjective Optimization Recent Trends in Evolutionary Multiobjective Optimization Self-adaptation and Convergence of Multiobjective Evolutionary Algorithms in Continuous Search Spaces A simple approach to evolutionary multi-objective optimization Quad-trees: A Data Structure for Storing Pareto-sets in Multi-objective Evolutionary Algorithms with Elitism Scalable Test Problems for Evolutionary Multi-Objective Optimization Particle Swarm Inspired Evolutionary Algorithm (PS-EA) for Multi-Criteria Optimization Problems Evolving Continuous Pareto Regions MOGADES: Multi-Objective Genetic Algorithm with Distributed Environment Scheme Use of Multiobjective Optimization Concepts to Handle Constraints in Genetic Algorithms Multi- Criteria Optimization of Finite State Automata: Maximizing Performance while Minimizing Description Length Multi-objective Optimization of Space Structures under Static and Seismic Loading Conditions

Bioinformatics using computational intelligence paradigms

Abstract: Medical Bioinformatics: Detecting Molecular Diseases with Case-Based Reasoning.- Prototype Based Recognition of Splice Sites.- Contact Based Image Compression in Biomedical High-Throughput Screening Using Artificial Neural Networks.- Discriminative Clustering of Yeast Stress Response.- A Dynamic Model of Gene Regulatory Networks Based on Inertia Principle.- Class Prediction with Microarray Datasets.- Random Voronoi Ensembles for Gene Selection in DNA Microarray Data.- Cancer Classification with Microarray Data Using Support Vector Machines.- Artificial Neural Networks for Reducing the Dimensionality of Gene Expression Data.

Cognitive Engineering: A Distributed Approach to Machine Intelligence

Abstract: The Psychological Basis of Cognitive Modeling.- Parallel and Distributed Logic Programming.- Distributed Reasoning by Fuzzy Petri Nets: A Review.- Belief Propagation and Belief Revision Models in Fuzzy Petri Nets.- Building Expert Systems Using Fuzzy Petri Nets.- Distributed Learning Using Fuzzy Cognitive Maps.- Unsupervised Learning by Fuzzy Petri Nets.- Supervised Learning by a Fuzzy Petri Net.- Distributed Modeling of Abduction, Reciprocity, and Duality by Fuzzy Petri Nets.- Human Mood Detection and Control: A Cybernetic Approach.- Distributed Planning and Multi-agent Coordination of Robots.

Beyond trust: a belief-desire-intention model of confidence in an agent's intentions

Abstract: Trust plays a fundamental role in multi-agent systems in which tasks are delegated or agents must rely on others to perform actions that they themselves cannot do. The concept of trust may be generalised and considered as a level of confidence in one's predictions of another agent's future behaviour. This has applicability beyond that normally ascribed to trust: for instance, one may be confident that a particular agent's intentions are hostile, and that this will be borne out by particular behaviours. In this paper we present a cognitive model of trust in which the central component is a Belief-Desire-Intention model or 'theory of mind' of a person or agent that evolves over time.

Design of intelligent multi-agent systems : human-centredness, architectures, learnings and adaptation

Abstract: 1. Humanization of soft computing agents.- 2. Software agents for ubiquitous computing.- 3. Agents-based knowledge logistics.- 4. Architectural styles and patterns for multi-agent systems.- 5. Design and behavior of a massive organization of agents.- 6. Developing agent-based applications with JADE.- 7. A collective can do better.- 8. Coordinating multi-agent assistants with an application by means of computational reflection.- 9. Learning by exchanging advice.- 10. Adaptation and mutation in multi-agent systems and beyond.- 11. Intelligent action acquisition for animated learning agents.- 12. Using stationary and mobile agents for information retrieval and e-commerce.

Advanced techniques in data mining and knowledge discovery

Abstract: Trends in Data Mining and Knowledge Discovery Advanced Methods for the Analysis of Semiconductor Manufacturing Process Data Clustering and visualization of Retail Market Baskets Segmentation Of Continuous Data Streams Based on a Change Detection Methodology Instance Selection Using Evolutionary Algorithms: An Experimental Study Using Cooperative Coevolution for Data Mining of Bayesian Networks Knowledge Discovery and Data Mining in Medicine Satellite Image Classification Using Cascaded Architecture of Neural Fuzzy Network Discovery of Positive and Negative Rules from Medical Databases based on Rough Sets

Practical applications of neural networks

Abstract: Neural networks have been applied with tremendous success to a wide range of real-world problems. The authors are continuing to be delighted to present the advantages offered by various neural networks in developing applications of great practical value where other techniques employed have failed. Neural networks have proven their practical applicability to almost any type of research or application area where learning for further recognition, prediction or classification tasks is concerned. This special issue comprises six papers selected from more than 350 presented at the Seventh International Conference on Knowledge-Based Intelligent Information and Engineering Systems (KES2003), 3–5 September 2003, University of Oxford, UK. The annual KES Conference has become a traditional and prestigious world forum for the presentation of developments relating to a wide spectrum of intelligent techniques and applications, with neural networks being among the most popular topics. The selected papers showcase a range of applications of neural networks to real-world problems, e.g. bioinformatics, medical diagnosis, fault diagnosis and image and pattern recognition. In addition to an interesting range of applications, the papers included in this issue come up with new techniques and approaches on hot topics in the area of neural networks, such as quantum neural networks, neural-network-based immune and multi-agent systems and multi-classifier systems. Today human–computer interfaces require advanced image processing systems. One difficult task in image processing is the localisation of the human face in an image. In the first paper of this issue, S. Behnke from the University of Freiburg, Germany proposed a hierarchical neural network architecture with local recurrent connectivity for face localisation and tracking. The proposed network is able to recognise human faces in the presence of noise, complex backgrounds or difficult lighting. Moreover, the network can be used for realtime face tracking, e.g. for accurately localising human faces in moving images. A delicate issue in machine learning is how to deal with incomplete datasets. C.P. Lim, M.M. Kuan and R.F. Harrison present a FAM-FCM hybrid neural network model for pattern classification when incomplete data sets are available. The fuzzy ARTMAP (FAM) network, which integrates fuzzy techniques with adaptive resonance theory (ART) neural networks, is the main part of the system and is employed to classify the input samples under various conditions. The fuzzy cmeans (FCM) based clustering strategies involved in this research are responsible for handling samples with missing features. The authors demonstrate the practical applicability of the FAM-FCM hybrid system in a medical diagnosis problem where, usually, the datasets available for making the diagnosis are incomplete. Over the years, researchers in the field of neural networks have proposed various types of neural networks, and have shown their limitations in a variety of real-world applications. Recently, quantum computing has been seen as a good solution for improving the abilities of neural networks. The third paper, by N. Kouka, N. Matsui, H. Nishimura and F. Pepper, introduces a quantum-computing-based neuron model, called Qubit. The quantum neural networks built using Qubit are compared with classical feed-forward neural networks, as well as with complex neural networks, in some pattern recognition problems. The performance of their quantum neural networks is reported to be better. The fourth paper is concerned with an application of neural networks to the fascinating field of bioinformatics. R. Ranawana and V. Palade from the University of Oxford, UK present a neural network multi-classifier system for gene recognition in DNA sequences of the Escherichia Coli (E. Coli) microorganism. Using various encoding techniques for input data, more neural network classifiers are trained using the same dataset, and V. Palade (&) Oxford University, UK

Reasoning about time in a BDI architecture

Abstract: The philosophical roots of the Belief-Desire-Intention model lie in Bratman's formulation of an intention theory of planning, in which he sought to make sense of the notion of future-directed intention. Implementations of BDI mainly follow the original Procedural Reasoning System model. BDI has a sound logical basis, exemplified by the Logic Of Rational Agents. While the LORA formulation has a temporal logic component, however, this does not translate into any ability for the agent to reason about actual time. Being able to reason about actual time would bring significant benefits for BDI agents, such as the ability for agents to communicate deadlines and to plan and schedule activities in a cooperating group. Given a suitable representation of temporal knowledge, an agent could learn about the temporal aspects of its own actions and processes, and this knowledge could be used as input to the planning process. This paper outlines a possible implementation strategy for the representation of, and the capacity to reason about, actual time.

A vision system for partially occluded landmark recognition

Abstract: This paper describes a vision system for extracting and recognising partially occluded 2D visual landmarks. The system is developed based on the traditional template matching approach and a memory feedback modulation (MFM) mechanism. It identifies the obscured portions and selectively enhances non-occluded areas of the landmark, while simultaneously suppressing background clutters of the bottom-up edge processed input images. The architecture has been tested with a large number of real images with varying levels of landmark concealment and further evaluated using a vision-based navigating robot in the laboratory environment.

Intelligent Software Systems in Historical Context

Abstract: automated reasoning capabilities 1 . It is argued that a distinction may be drawn between h uman intelligence and component capabilities within a more genera l definition of intelligence, and that such component capabilities can be embedded in computer software. The primary vehicle in the quest for intelligent software has been the gradual recognition of the central role played by data and information, rather than the logic and functionality of the applic ation. The three milestones in this evolution have been: the separation of data management from the inter nal domain of the application; the development of standard data exchange protocols such as the Extensible Markup Language (XML), which allows machine interpretable structure and meaning to be added to data exchange packages; and, the ability to build information models that are rich in relationships and are thereby capable of supporting the automated reasoning capabilities of software agents. To assess the relative capabilities of software system s during their evolutionary path over the past 50 years the author utilizes an evaluation framework cons isting of six categories, with appropriate features or capabilities under each category serving as a set of e valuation criteria. Designed to assess the degree to which a software application or system is capable of performing intelligent functions, the particular

Shadow watermark extraction system

Abstract: A new watermarking scheme which provides the ability of sharing secret with multi-users is proposed It splits the original watermark into two shares and embeds one share into the cover image to increase the security A polarization procedure is performed to establish a polarity stream from the cover image The second share and the polarity stream are used to generate a master key and several normal keys In our system, only the super user can reveal the genuine watermark directly Other users possess the normal keys can obtain shadow watermarks merely By combining the shadow watermarks together, the real watermark can be recovered

Collaborative agent learning using hybrid neurocomputing.

Abstract: This paper investigates the use of a hybrid neurocomputing approach to detect and then recognise images. The first technique creates and trains intelligent agents capable of detecting face images by using a Generalised Regression Neural Network (GRNN). The second technique further refines the search by recognising images from the detected data set using a feed forward backpropagation neural network. These two agents make up the ‘Detection Agent’ and the ‘Recognition Agent’ in an agent architecture that collaborates with each other to detect and then recognise certain images. The overall agent architecture will operate as an ‘Automatic Target Recognition’ (ATR) system. The architecture of ATR system is presented in this paper and it is shown how the Detection and Recognition Agents (DRA) fit into the overall system. Experiments and results using the DRA are also presented.

Learning Algorithm of Neural Network using Orthogonal Decomposition Method

Abstract: In this paper, we present a new learning algorithm of neural network based on the orthogonal decomposition method (ORT). The main scheme of this algorithm is using the ORT to obtain specially structured subspaces defined by the input-output data. This structure is then exploited in the calculation of the parameter estimation of the neural network. Therefore, the method to obtain the comparatively accurate estimate is introduced without iteration calculations. We show that this algorithm can be applied to successfully identify the nonlinear system in the presence of comparatively loud noise. Results from several simulation studies have been included to the effectiveness of this method.