We propose a new framework to construct the margin-based classifiers, in which the binary and multicategory classification problems are solved by the same principle; namely, margin-based classification via regularized empirical risk minimization. To build the framework, we propose the margin vector which is the multi-class generalization of the margin, then we further generalize the concept of ...

Records from two connected proglacial threshold lakes at Paakitsoq, west Greenland have been analyzed in order to investigate the response of a land-based ice sheet margin to Holocene climate change. The results are used to test whether or not the land terminating margin at Paakitsoq behaved synchronously with the nearby marine terminating Jakobshavn Isbræ during the Holocene. The radiocarbon d...

Abstract In this paper, first, the stability margin and stabilizability margin of multidimensional (nD) systems are considered. In particular, the concept of stabilizability margin is introduced for the first time and it is defined to be the largest stability margin that a closed-loop feedback system can reach by menas of any stabilizing compensator. The investigation of stabilizability margin ...

Finding linear classifiers that maximize AUC scores is important in ranking research. This is naturally formulated as a 1-norm hard/soft margin optimization problem over pn pairs of p positive and n negative instances. However, directly solving the optimization problems is impractical since the problem size (pn) is quadratically larger than the given sample size (p + n). In this paper, we give ...