K
Kah-Kay Sung
Researcher at National University of Singapore
Publications - 4
Citations - 1483
Kah-Kay Sung is an academic researcher from National University of Singapore. The author has contributed to research in topics: Heuristic & Stability (learning theory). The author has an hindex of 2, co-authored 4 publications receiving 1376 citations. Previous affiliations of Kah-Kay Sung include Massachusetts Institute of Technology.
Papers
More filters
Journal ArticleDOI
Comparing support vector machines with Gaussian kernels to radial basis function classifiers
Bernhard Schölkopf,Kah-Kay Sung,C.J.C. Burges,Federico Girosi,Partha Niyogi,Tomaso Poggio,Vladimir Vapnik +6 more
TL;DR: The results show that on the United States postal service database of handwritten digits, the SV machine achieves the highest recognition accuracy, followed by the hybrid system, and the SV approach is thus not only theoretically well-founded but also superior in a practical application.
Patent
Network-based system and method for detection of faces and the like
Tomaso Poggio,Kah-Kay Sung +1 more
TL;DR: In this article, a face detection system (100) includes an imaging device, a computer having a pattern prototype synthetizer and an image classifier, and an output display device, which synthesizes face and non-face pattern prototypes by a network training process using a number of example images.
Book ChapterDOI
An active learning formulation for instance selection with applications to object detection
Kah-Kay Sung,Partha Niyogi +1 more
TL;DR: This chapter presents a Bayesian formulation for active learning within a classical function approximation learning framework, and shows how one can derive precise example selection algorithms for learning some simple target function classes more accurately with less training data.
Journal ArticleDOI
Epsilon focusing—a strategy for active example selection
Partha Niyogi,Kah-Kay Sung +1 more
TL;DR: An e-focusing strategy that actively chooses examples for concept learning is discussed, and the local focused property that functions must have for such a strategy to work is described.