Sciweavers

3643 search results - page 530 / 729
» Learning Submodular Functions
Sort
View
ICML
2007
IEEE
16 years 7 months ago
Asymmetric boosting
A cost-sensitive extension of boosting, denoted as asymmetric boosting, is presented. Unlike previous proposals, the new algorithm is derived from sound decision-theoretic princip...
Hamed Masnadi-Shirazi, Nuno Vasconcelos
ICML
2008
IEEE
16 years 7 months ago
Robust matching and recognition using context-dependent kernels
The success of kernel methods including support vector machines (SVMs) strongly depends on the design of appropriate kernels. While initially kernels were designed in order to han...
Hichem Sahbi, Jean-Yves Audibert, Jaonary Rabariso...
ICML
2008
IEEE
16 years 7 months ago
Efficiently solving convex relaxations for MAP estimation
The problem of obtaining the maximum a posteriori (map) estimate of a discrete random field is of fundamental importance in many areas of Computer Science. In this work, we build ...
M. Pawan Kumar, Philip H. S. Torr
ICML
2006
IEEE
16 years 7 months ago
Algorithms for portfolio management based on the Newton method
We experimentally study on-line investment algorithms first proposed by Agarwal and Hazan and extended by Hazan et al. which achieve almost the same wealth as the best constant-re...
Amit Agarwal, Elad Hazan, Satyen Kale, Robert E. S...
ICML
2006
IEEE
16 years 7 months ago
Personalized handwriting recognition via biased regularization
We present a new approach to personalized handwriting recognition. The problem, also known as writer adaptation, consists of converting a generic (user-independent) recognizer int...
Wolf Kienzle, Kumar Chellapilla