Kar-Ann Toh

Co-authors

Following

Followers

Research Focus & Metrics

Explore full metrics on Scilit

Academic Career Length

21Years (2006-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

1 June 2008 in Neural Computation

Journal Article

Deterministic Neural Classification

Kar-Ann Toh

This letter presents a minimum classification error learning formulation for a single-layer feedforward network (SLFN). By approximating the nonlinear counting step function using a quadratic function, the classification error rate is shown to be deterministically solvable. Essentially the derived solution is related to an existing weighted least-squares method with class-specific weights set according to the size of data set. By considering the class-specific weights as adjustable parameters, the learning formulation extends the classification robustness of the SLFN without sacrificing its intrinsic advantage of being a closed-form algorithm. While the method is applicable to other linear formulations, our empirical results indicate SLFN's effectiveness on classification generalization.

4 April 2008 in IEEE Transactions on Pattern Analysis and Machine Intelligence

Journal Article

Between Classification-Error Approximation and Weighted Least-Squares Learning

Kar-Ann Toh

How-Lung Eng

This paper presents a deterministic solution to an approximated classification-error-based objective function. In the formulation, we propose a quadratic approximation as the function for achieving smooth error counting. The solution is subsequently found to be related to the weighted least-squares, whereby a robust tuning process can be incorporated. The tuning traverses between the least- squares estimate and the approximated total-error-rate estimate to cater to various situations of unbalanced attribute distributions. By adopting a linear parametric classifier model, the proposed classification-error-based learning formulation is empirically shown to be superior to that using the original least-squares-error cost function. Finally, it will be seen that the performance of the proposed formulation is comparable to other classification-error-based and state-of-the-art classifiers without sacrificing the computational simplicity.

7 August 2007 in Pattern Recognition

Journal Article

Biometric scores fusion based on total error rate minimization

Kar-Ann Toh

Jaihie Kim

Sangyoun Lee

18 March 2021 in IEEE Transactions on Systems, Man, and Cybernetics: Systems

Journal Article

Blockwise Recursive Moore–Penrose Inverse for Network Learning

Huiping Zhuang

Zhiping Lin

Kar-Ann Toh

Training neural networks with the Moore–Penrose (MP) inverse has recently gained attention in view of its noniterative training nature. However, a significant drawback of learning based on the MP inverse is that the computational memory consumption grows along with the size of a dataset. In this article, based on the partitioning of the MP inverse, we propose a blockwise recursive MP inverse formulation (BRMP) for network learning with low-memory property while preserving its training effectiveness. The BRMP is an equivalent formulation to its batchwise counterpart since neither approximation nor assumption is made in the derivation process. Our further exploration of this recursive method leads to a switching structure among three different scenarios. This structure also reveals that the well-known recursive least squares method is a special case of our proposed technique. Subsequently, we apply BRMP to the training of radial basis function networks as well as multilayer perceptrons. The experimental validation covers both regression and classification tasks.

1 September 2006

Conference Paper

Changeable Biometrics for Appearance Based Face Recognition

To enhance security and privacy in biometrics, changeable (or cancelable) biometrics have recently been introduced. The idea is to transform a biometric signal or feature into a new one for enrollment and matching. In this paper, we proposed changeable biometrics for face recognition using an appearance based approach. PCA and ICA coefficient vectors extracted from an input face image are normalized using their norm. The two normalized vectors are scrambled randomly and a new transformed face coefficient vector (transformed template) is generated by addition of the two normalized vectors. When a transformed template is compromised, it is replaced by using a new scrambling rule. Because the transformed template is generated by the addition of two vectors, the original PCA and ICA coefficients cannot be recovered from the transformed coefficients. In our experiment, we compared the performance between the cases when PCA and ICA coefficient vectors are used for verification and when the transformed coefficient vectors are used for verification.

30 December 2019 in IEEE Access

Journal Article

Extraction and Cross-Matching of Palm-Vein and Palmprint From the RGB and the NIR Spectrums for Identity Verification

In this paper, we propose a novel cross-spectral matching system for identity verification based on the palm-vein and the palmprint acquired from the visible (RGB) and the near infrared (NIR) image spectral bands. Considering the vast availability of the visible library, the red and the blue spectrums are treated as sources of gallery samples and the NIR spectral band is utilized as the probe source without loss of generality. Apart from the extraction of palm-vein and palmprint features, the discriminative power of the palmprint templates is enhanced using a simplified Local Binary Pattern (LBP) encoding scheme. The similarity scores obtained by matching the NIR palm-vein templates against the registered RGB palm-vein templates is finally fused with scores obtained from matching the NIR palmprint codes against the registered RGB palmprint codes. Our empirical results on two publicly available multi-spectral palm databases show that the proposed system consistently achieves promising verification performance.

About Me

Prof. Kar Ann Toh is a professor at the School of Electrical and Electronic Engineering at Yonsei University, South Korea. He received a PhD degree from Nanyang Technological University (NTU), Singapore, in 1999. He worked for two years in the aerospace industry before his post-doctoral appointments at research centers in NTU from 1998 to 2002. He was affiliated with the Institute for Infocomm Research in Singapore from 2002 to 2005 before his current appointment in Korea. During his sabbatical year in 2020, he worked with the Department of Electrical and Computer Engineering, National University of Singapore (ECE-NUS), as a visiting professor. Prof. Kar Ann Toh's research interests include biometrics, pattern classification, optimization, and neural networks. He is a co-inventor of a US patent and has made several PCT filings related to biometric applications. In addition to being an active member of publications (PAMI, Machine Learning, Neural Computation, etc.), Prof. Kar Ann Toh has served on the technical program committee for international conferences related to biometrics and artificial intelligence. He has also served as a reviewer for international journals including several IEEE Transactions. He is a senior member of the IEEE.

Kar-Ann Toh

Kar-Ann Toh

Research Focus & Metrics

Academic Career Length

Publication Type Breakdown

Representative Publications

About Me