Maximing the Margin in the Input Space
نویسنده
چکیده
We propose a novel criterion for support vector machine learning: maximizing the margin in the input space, not in the feature (Hilbert) space. This criterion is a discriminative version of the principal curve proposed by Hastie et al. The criterion is appropriate in particular when the input space is already a well-designed feature space with rather small dimensionality. The definition of the margin is generalized in order to represent prior knowledge. The derived algorithm consists of two alternating steps to estimate the dual parameters. Firstly, the parameters are initialized by the original SVM. Then one set of parameters is updated by Newton-like procedure, and the other set is updated by solving a quadratic programming problem. The algorithm converges in a few steps to a local optimum under mild conditions and it preserves the sparsity of support vectors. Although the complexity to calculate temporal variables increases the complexity to solve the quadratic programming problem for each step does not change. It is also shown that the original SVM can be seen as a special case. We further derive a simplified algorithm which enables us to use the existing code for the original SVM.
منابع مشابه
The Design of a New Double Loop Controller For Simultaneous Adjustment of Input and Output Voltages of Single-Phase Grid-Connected Inverter
Although LCL filters are used widely in the grid connected inverters to reduce high-orderharmonics, such a system increases system order and therefore sustainable design of closed-loopcontroller system will be complicated. Recently, the single-loop control strategy has beensuggested for L or LC filter based grid-connected inverters. However, the use of single-loopcontrol directly in LCL filter-...
متن کاملEnhancing Efficiency of Neural Network Model in Prediction of Firms Financial Crisis Using Input Space Dimension Reduction Techniques
The main focus in this study is on data pre-processing, reduction in number of inputs or input space size reduction the purpose of which is the justified generalization of data set in smaller dimensions without losing the most significant data. In case the input space is large, the most important input variables can be identified from which insignificant variables are eliminated, or a variable ...
متن کاملInvestigating the Effect of Variations in Irrigation Water Price on Cropping Pattern and Gross Margin under Uncertainty (Case Study: Khorasan Razavi)
Water shortage crisis is an issue that has led to drastic changes in different agricultural policies, especially in arid and semi-arid areas. Uncertainty in the amount of resources, e.g. water, used for agricultural production entails risk for farmers' income and cropping pattern changes. In the present study, the robust optimization model was used for optimal allocation of arable lands of Khor...
متن کاملOrganizing the spatial space of the urban rivers' margin of tourism development (case study of Poldokhtar city)
This study aims at organizing the city of Poldokhtar in the city of Kashkan. Using this element of natural value, the city is being restored and organized, and its potential for tourism and development of tourism and economic prosperity is to be used. The paths of the rivers within the city can be used as a linear route, and can be used as a sustainable urban tourism axis. In the case of the sp...
متن کاملMarginal Assessment of Crowns by the Aid of Parallel Radiography
Introduction: Marginal adaptation is the most critical item in long-term prognosis of single crowns. This study aimed to assess the marginal quality as well asthe discrepancies in marginal integrity of some PFM single crowns of posterior teeth by employing parallel radiography in Shiraz Dental School, Shiraz, Iran. Methods: In this descriptive study, parallel radiographies were taken from 200 f...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره cs.AI/0211006 شماره
صفحات -
تاریخ انتشار 2002