A. Mohan

Followers

Following

Co-authors

Public Views

Interests

Uploads

Papers by A. Mohan

Genetic algorithms for graph partitioning and incremental graph partitioning

Conference on High Performance Computing (Supercomputing), Nov 14, 1994

Abstract Partitioning graphs into equally large groups of nodes, minimizing the number of edges b... more

Multi-Objective Evolutionary Algorithms for Sensor Network Design

IGI Global eBooks, May 24, 2011

Many sensor network design problems are characterized by the need to optimize multiple conflictin... more Many sensor network design problems are characterized by the need to optimize multiple conflicting objectives. However, existing approaches generally focus on a single objective (ignoring the others), or combine multiple objectives into a single function to be optimized, to facilitate the application of classical optimization algorithms. This restricts their ability and constrains their usefulness to the network designer. A much more appropriate and natural approach is to address multiple objectives simultaneously, applying recently developed multi-objective evolutionary algorithms (MOEAs) in solving sensor network design problems. This chapter describes and illustrates this approach by modeling two sensor network design problems (mobile agent routing and sensor placement), as multi-objective optimization problems, developing the appropriate objective functions and discussing the tradeoffs between them.

Download

Evolution of Space-Partitioning Forest for Anomaly Detection

Previous work proposed a fast one-class anomaly detector using an ensemble of random half-space p... more Previous work proposed a fast one-class anomaly detector using an ensemble of random half-space partitioning trees. The method was shown to be effective and efficient for detecting anomalies in streaming data. However, the parameters were pre-defined, so the random partitions of the data space might not be optimal. Therefore, the aims of this study were to: (a) give some mathematical analysis of the random partitioning trees; and (b) explore optimizing forests for anomaly detection using evolutionary algorithms.

Detection of anomalous time series based on multiple distance measures

Structured motif extraction using affinity based cluster analysis

Optimal bandwidth assignment for distributed sequential detection

Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)

ABSTRACT

Ensemble Algorithms for Unsupervised Anomaly Detection

Lecture Notes in Computer Science, 2015

Many anomaly detection algorithms have been proposed in recent years, including density-based and... more Many anomaly detection algorithms have been proposed in recent years, including density-based and rank-based algorithms. In this paper, we propose ensemble methods to improve the performance of these individual algorithms. We evaluate approaches that use score and rank aggregation for these algorithms. We also consider sequential methods in which one detection method is followed by the other. We use several datasets to evaluate the performance of the proposed ensemble methods. Our results show that sequential methods significantly improve the ability to detect anomalous data points.

A neural network simulator for the Connection Machine

Proceedings. 5th IEEE International Symposium on Intelligent Control 1990

In this paper we describe the design, development, and performance of a neural network simulator ... more In this paper we describe the design, development, and performance of a neural network simulator for the Connection Machine (CM) 3 • The design of the simulator is based on the Rochester Connectionist Simulator(RCS). RCS is a simulator for connectionist networks developed at the University of Rochester. The CM simulator can be used as a stand-alone system or as a high-performance parallel back-end to RCS. In the latter case, once the network has been built by RCS, the high-performance parallel back-end system constructs an equivalent network on the CM processor array and executes it. The CM simulator facilitates the exploitation of the massive parallelism inherent in connectionist networks. It can also enable substantial reduction in the training times of connectionist networks. 3 Connection Machine is a registered trademark of Thinking Machines Corporation.

Download

SMAlign: Alignment of DNA Sequences with Gap Constraints

Coupling of airflow and pollutant dispersion models with evacuation planning algorithms for building system controls

A Network Theoretic Analysis of Evolutionary Algorithms

Lecture Notes in Computer Science, 2012

ABSTRACT Network theoretic analyses have been shown to be extremely useful in multiple fields and... more ABSTRACT Network theoretic analyses have been shown to be extremely useful in multiple fields and applications. We propose this approach to study the dynamic behavior of evolutionary algorithms, the first such analysis to the best of our knowledge. Evolving populations are represented as dynamic networks, and we show that changes in population characteristics can be recognized at the level of the networks representing successive generations, with implications for possible improvements in the evolutionary algorithm, e.g., in deciding when a population is prematurely converging, and when a reinitialization of the population may be beneficial to reduce computational effort. In this paper, we show that network-theoretic analyses of evolutionary algorithms help in: (i) studying community-level behaviors, and (ii) using graph properties and metrics to analyze evolutionary algorithms.

Perturbed-attractor particle swarm optimization for image restoration

Institution of Engineering and Technology eBooks, Oct 9, 2018

This chapter addresses the problem of restoration of images degraded by blurring and noise. Image... more This chapter addresses the problem of restoration of images degraded by blurring and noise. Image degradation is modeled as the convolution of a low-pass, two-dimensional point spread function with a reference or undegraded image. The image restoration problem is modeled as an optimization problem and a modified particle swarm optimization algorithm is developed. This approach is evaluated using image quality metrics that consider the quality of edges restored as well as the fidelity of the restored image with respect to the undegraded reference image. A novel-edge sharpness metric is also proposed. The new approach is shown to perform well in several image restoration examples.

Linkage crossover operator for genetic algorithms

Problem-specific knowledge is often implemented in search algorithms using heuristics to determin... more Problem-specific knowledge is often implemented in search algorithms using heuristics to determine which search paths are to be explored at any given instant. As in other search methods, utilizing this knowledge will lead a genetic algorithm (GA) faster towards better results. In many problems, crucial knowledge is to be found not in individual components, but in interrelations between those components. For such problems, we develop an interrelation (linkage) based crossover operator that has the advantage of liberating GAs from the constraints imposed by the fixed representations generally chosen for problems. The strengths of linkages between components of a chromosomal structure can be explicitly represented in a linkage matrix and used in the reproduction step to generate new individuals. For some problems, such a linkage matrix is known a priori from the nature of the problem. In other cases, the linkage matrix may be learned by successive minor adaptations during the execution of the evolutionary algorithm. We present four linkage adaptation procedures that work along with the linkage based crossover. We demonstrate the success of such an approach for several benchmark problems representing two types of problems: One where the linkage structure is well-defined and is easily represented as linkage matrix in advance, and another where the linkage structure is not well-known. Linkage adaptation procedures show success in learning the right linkage structure for a given problem, and obtaining better results than the standard crossover operators. This approach is shown to be robust to different variations of GA parameter values

Linkage crossover for genetic algorithms

A new linkage crossover" operator based on probabilistic methodology is proposed. The genetic alg... more A new linkage crossover" operator based on probabilistic methodology is proposed. The genetic algorithm using this operator is shown to be particularly successful for deceptive problems with linkages that are not easily captured by linear relationships. Further, adaptive Hebbian mechanisms are shown to be successful in learning the appropriate linkage relationships when faced with new problems. Results demonstrate the e cacy of this approach, compared with traditional general purpose crossover operators.

Download

A neural network approach to prediction of psychological performance in frequency discrimination

1990 IJCNN International Joint Conference on Neural Networks, 1990

ABSTRACT

Training techniques to obtain fault-tolerant neural networks

Proceedings of IEEE 24th International Symposium on Fault- Tolerant Computing

This paper addresses methods of improving the faulttolerance of feedforward neural nets. The firs... more This paper addresses methods of improving the faulttolerance of feedforward neural nets. The first methodis to coerce weights to have low magnitudes during thebackpropagation training process, since fault toleranceis degraded by the use of high magnitude weights; at thesame time, additional hidden nodes are added dynamicallyto the network to ensure that desired performancecan be obtained. The second method is to

Elements of Artificial Neural Network

Modifying training algorithms for improved fault tolerance

Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94), 1994

ABSTRACT

Acyclic Identification of Aptamer from Over-Represented Libraries Using Hash Functions

2013 39th Annual Northeast Bioengineering Conference, 2013

In recent years, with the advent of fast sequencing technology, the genomic database is growing r... more In recent years, with the advent of fast sequencing technology, the genomic database is growing rapidly. Researchers in bioinformatics field are expecting faster and more accurate tools to effectively analyze the gigantic data sets. In the context of aptamer search, the goal is to search for the over-represented DNA sequences compared with random background libraries on the same chip. Hash functions are widely used in substring comparison, sequence alignment and clustering tools. We have developed a light-weighted tool that takes advantage of the hash functions to reduce the size of genomic data and conducts k-neighbor searches on the centroid sequence. This greatly improves the efficiency of the search compared with the existing tool. Furthermore, the calculation of k-neighbor hash values decreases the mutant searching overhead. In a dataset of 1 million sequences, the program accurately counted the frequency of the Human alpha-Thrombin sequence and found the mutant versions of the target sequence in less than 40 seconds, whereas the existing method takes 8280 seconds (2 hours 13 minutes).

An Online Anomalous Time Series Detection Algorithm for Univariate Data Streams

Lecture Notes in Computer Science, 2013

ABSTRACT We address the online anomalous time series detection problem among a set of series, com... more ABSTRACT We address the online anomalous time series detection problem among a set of series, combining three simple distance measures. This approach, akin to control charts, makes it easy to determine when a series begins to differ from other series. Empirical evidence shows that this novel online anomalous time series detection algorithm performs very well, while being efficient in terms of time complexity, when compared to approaches previously discussed in the literature.

A. Mohan

Uploads

Papers by A. Mohan

Log In