Results 1  10
of
1,975
Feature selection based on mutual information: Criteria of maxdependency, maxrelevance, and minredundancy
 IEEE TRANS. PATTERN ANALYSIS AND MACHINE INTELLIGENCE
, 2005
"... Feature selection is an important problem for pattern classification systems. We study how to select good features according to the maximal statistical dependency criterion based on mutual information. Because of the difficulty in directly implementing the maximal dependency condition, we first der ..."
Abstract

Cited by 571 (8 self)
 Add to MetaCart
derive an equivalent form, called minimalredundancymaximalrelevance criterion (mRMR), for firstorder incremental feature selection. Then, we present a twostage feature selection algorithm by combining mRMR and other more sophisticated feature selectors (e.g., wrappers). This allows us to select a
An Improved Training Algorithm for Support Vector Machines
, 1997
"... We investigate the problem of training a Support Vector Machine (SVM) [1, 2, 7] on a very large date base (e.g. 50,000 data points) in the case in which the number of support vectors is also very large (e.g. 40,000). Training a SVM is equivalent to solving a linearly constrained quadratic programmin ..."
Abstract

Cited by 339 (1 self)
 Add to MetaCart
We investigate the problem of training a Support Vector Machine (SVM) [1, 2, 7] on a very large date base (e.g. 50,000 data points) in the case in which the number of support vectors is also very large (e.g. 40,000). Training a SVM is equivalent to solving a linearly constrained quadratic
The Octagon Abstract Domain
, 2007
"... ... domain for static analysis by abstract interpretation. It extends a former numerical abstract domain based on DifferenceBound Matrices and allows us to represent invariants of the form (±x ± y ≤ c), where x and y are program variables and c is a real constant. We focus on giving an efficient re ..."
Abstract

Cited by 321 (24 self)
 Add to MetaCart
representation based on DifferenceBound Matrices—O(n 2) memory cost, where n is the number of variables—and graphbased algorithms for all common abstract operators—O(n 3) time cost. This includes a normal form algorithm to test equivalence of representation and a widening operator to compute least fixpoint
Scalable informationdriven sensor querying and routing for ad hoc heterogeneous sensor networks
 International Journal of High Performance Computing Applications
, 2002
"... This paper describes two novel techniques, informationdriven sensor querying (IDSQ) and constrained anisotropic diffusion routing (CADR), for energyefficient data querying and routing in ad hoc sensor networks for a range of collaborative signal processing tasks. The key idea is to introduce an inf ..."
Abstract

Cited by 277 (12 self)
 Add to MetaCart
an information utility measure to select which sensors to query and to dynamically guide data routing. This allows us to maximize information gain while minimizing detection latency and bandwidth consumption for tasks such as localization and tracking. Our simulation results have demonstrated
PATH PLANNING IN EXPANSIVE CONFIGURATION SPACES
, 1999
"... We introduce the notion of expansiveness to characterize a family of robot configuration spaces whose connectivity can be effectively captured by a roadmap of randomlysampled milestones. The analysis of expansive configuration spaces has inspired us to develop a new randomized planning algorithm. ..."
Abstract

Cited by 264 (30 self)
 Add to MetaCart
We introduce the notion of expansiveness to characterize a family of robot configuration spaces whose connectivity can be effectively captured by a roadmap of randomlysampled milestones. The analysis of expansive configuration spaces has inspired us to develop a new randomized planning algorithm
Logistic Regression, AdaBoost and Bregman Distances
, 2000
"... We give a unified account of boosting and logistic regression in which each learning problem is cast in terms of optimization of Bregman distances. The striking similarity of the two problems in this framework allows us to design and analyze algorithms for both simultaneously, and to easily adapt al ..."
Abstract

Cited by 259 (45 self)
 Add to MetaCart
We give a unified account of boosting and logistic regression in which each learning problem is cast in terms of optimization of Bregman distances. The striking similarity of the two problems in this framework allows us to design and analyze algorithms for both simultaneously, and to easily adapt
Containment and equivalence for a fragment of XPath
 JOURNAL OF THE ACM
, 2004
"... XPath is a language for navigating an XML document and selecting a set of element nodes. XPath expressions are used to query XML data, describe key constraints, express transformations, and reference elements in remote documents. This article studies the containment and equivalence problems for a ..."
Abstract

Cited by 142 (0 self)
 Add to MetaCart
XPath is a language for navigating an XML document and selecting a set of element nodes. XPath expressions are used to query XML data, describe key constraints, express transformations, and reference elements in remote documents. This article studies the containment and equivalence problems for a
Minimization of Tree Pattern Queries
 In SIGMOD
, 2001
"... Tree patterns form a natural basis to query treestructured data such as XML and LDAP. Since the efficiency of tree pattern matching against a treestructured database depends on the size of the pattern, it is essential to identify and eliminate redundant nodes in the pattern and do so as quickly as ..."
Abstract

Cited by 137 (4 self)
 Add to MetaCart
dependent minimization. For treestructured databases, required child/descendant and type cooccurrence ICs are very natural. Under such ICs, we show that the minimal equivalent query is unique. We show the surprising result that the algorithm obtained by first augmenting the tree pattern using ICs, and then applying CIM
Conjunctive Query Containment Revisited
, 1998
"... We consider the problems of conjunctive query containment and minimization, which are known to be NPcomplete, and show that these problems can be solved in polynomial time for the class of acyclic queries. We then generalize the notion of acyclicity and define a parameter called query width that ca ..."
Abstract

Cited by 117 (0 self)
 Add to MetaCart
that captures the "degree of cyclicity" of a query: in particular, a query is acyclic if and only if its query width is 1. We give algorithms for containment and minimization that run in time polynomial in n k , where n is the input size and k is the query width. These algorithms naturally
Efficient Search for Approximate Nearest Neighbor in High Dimensional Spaces
, 1998
"... We address the problem of designing data structures that allow efficient search for approximate nearest neighbors. More specifically, given a database consisting of a set of vectors in some high dimensional Euclidean space, we want to construct a spaceefficient data structure that would allow us to ..."
Abstract

Cited by 215 (9 self)
 Add to MetaCart
We address the problem of designing data structures that allow efficient search for approximate nearest neighbors. More specifically, given a database consisting of a set of vectors in some high dimensional Euclidean space, we want to construct a spaceefficient data structure that would allow us
Results 1  10
of
1,975