Browsing by Author "Wimalawarne, KADNK"

Now showing 1 - 3 of 3

item: Conference-Full-text
Efficient high performance computing framework for short rate models
(Computer Science & Engineering Society c/o Department of Computer Science and Engineering, University of Moratuwa., 2009-07) Dampahala, TP; Premadasa, HDDD; Ranasinghe, PWW; Weerasinghe, JNP; Wimalawarne, KADNK; Nanayakkara, V; Gunasinghe, UP
Many mathematical calculations in the field of computational finance consume a lot of time and resources for processing. Some of the Short rate models used in quantitative finance which have been taken into consideration in this paper have been optimized for performance within a cluster computing environment. The back-end cluster has been seamlessly integrated with an easy-to-use front-end which can be used by finance professionals who are not aware of the details of the computational and database cluster. Furthermore, many techniques that have been utilized to improve the efficiency of the models have also been described. This paper also describes the generalization of a High Performance Computing Cluster designed for One-factor Short rate models and how it can be used easily to be further extended for other mathematical models in quantitative finance. The ultimate objective is to come up with a generalized framework for quantitative finance.
item: Conference-Extended-Abstract
Face recognition using informative vector machines
(2007) Wimalawarne, KADNK; De Silva, CR
In the recent past kernel methods have been successfully applied to face recognition. We present a novel approach in frontal face recognition with informative vector machine, a sparse Gaussian process kernel classifier. Informative vector machine has the ability to provide more sparse solutions than the widely used kernel classifiers like support vector machines.
item: Thesis-Abstract
Face recognition using kernel classifiers
Wimalawarne, KADNK; De Silva, C
Face reorganization to be one of the biggest challengers to the machine learning community Over three decades of extensive research has been earned out in this field by many researchers. In spite of many face recognition methods developed. Research on novel methods are needed m fulfill need of modern application// In the recent past kernel methods have teen successful!) applied to FACE recognition. We present a novel approach in face recognition with informative vector a machine a sparse gauss ion tor process kernel classifier. Experiments with the ORL . face database shows that recognition accuracies both these algorithms to be comparable . but informative vector machine . we also found that using automatic relevance determination kernel which with informative vector machine pan ides .1 novel approach lo dimension reduction is feature space. Overall. both sparse solution and dimension reduction* with informative vector machine reduces the storage space and computational cost while achieving a recognition accuracy close to supports vector machines