We propose a new method for clustering based on finding maximum margin hyperplanes through data. By reformulating the problem in terms of the implied equivalence relation matrix, we can pose the problem as a convex integer program. Although this still yields a difficult computational problem, the hard-clustering constraints can be relaxed to a soft-clustering formulation which can be feasibly s...

Many machine learning applications such as in vision, biology and social networking deal with data in high dimensions. Feature selection is typically employed to select a subset of features which improves generalization accuracy as well as reduces the computational cost of learning the model. One of the criteria used for feature selection is to jointly minimize the redundancy and maximize the r...

In this paper we examine the quantitative effects of margin regulation on volatility in asset markets. We consider a general equilibrium infinite-horizon economy with heterogeneous agents and collateral constraints. There are two assets in the economy which can be used as collateral for short-term loans. For the first asset the margin requirement is exogenously regulated while the margin requir...

We propose a new feature extraction method called Margin Maximizing Discriminant Analysis (MMDA) which seeks to extract features suitable for classification tasks. MMDA is based on the principle that an ideal feature should convey the maximum information about the class labels and it should depend only on the geometry of the optimal decision boundary and not on those parts of the distribution o...

We present a method to simultaneously learn a mixture of mappings and large margin hyperplane classifier. This method learns useful mappings of the training data to improve classification accuracy. We first present a simple iterative algorithm that finds a greedy local solution and then derive a semidefinite relaxation to find an approximate global solution. This relaxation leads to the matrix ...

We present a novel approach to collaborative prediction, using low-norm instead of low-rank factorizations. The approach is inspired by, and has strong connections to, large-margin linear discrimination. We show how to learn low-norm factorizations by solving a semi-definite program, and present generalization error bounds based on analyzing the Rademacher complexity of low-norm factorizations.

the biologic width (bw) includes attached epithelial cells and connective tissue attachment complex being very important in the periodontal health during prosthetic treatments as invading this zone can cause bone resorption and gingival recession. the present study investigated biologic width values in the normal periodontium in anterior and posterior teeth. 30 patients that referred from resto...

the study specifically described the socio-economic status of the people involved in the production, distribution and consumption of convenience foods in lafia urban of nasarawa state. it identified the factors that influence the entry into convenience food enterprise, factors necessitating the demand and supply of the products and examined the costs and returns of three convenience foods. the ...

Rectal cancer carries poor prognosis because of metastasis and local recurrence. Local recurrence has a profound effect on morbidity and quality of life. Therefore, preoperative staging of rectal cancer has an important impact on treatment plan. The main factor in predicting the local recurrence is the circumferential resection margin (CRM). Surgical resection with stage-appropriate neoadjuvant...

Mesoscopic simulations of long polymer chains and soft matter systems are conducted routinely in the literature order to assess long-lived relaxation processes manifested these systems. Coarse-grained are, however, prone unphysical intercrossing due their inherent softness. This issue can be resolved by introducing intermolecular bonds (the so-called slip-springs) which restore topological cons...