Gradient projection methods for quadratic programs and applications in training support vector machines

DOI10.1080/10556780512331318182zbMath1072.90026OpenAlexW2031127394MaRDI QIDQ5317754

Thomas Serafini, Gaetano Zanghirati, Luca Zanni

Publication date: 21 September 2005

Published in: Optimization Methods and Software (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1080/10556780512331318182

zbMATH Keywords

support vector machineslarge-scale problemsdecomposition techniquesquadratic programsgradient projection methods

Mathematics Subject Classification ID

Quadratic programming (90C20) Methods of reduced gradient type (90C52)

Related Items

Stable equilibrium configuration of two bar truss by an efficient nonmonotone global Barzilai-Borwein gradient method in a fuzzy environment, Binary separation and training support vector machines, Accelerating gradient projection methods for \(\ell _1\)-constrained signal recovery by steplength selection rules, Hybrid spectral gradient method for the unconstrained minimization problem, LMBOPT: a limited memory method for bound-constrained optimization, Partial spectral projected gradient method with active-set strategy for linearly constrained optimization, Non-negatively constrained image deblurring with an inexact interior point method, Comparison of active-set and gradient projection-based algorithms for box-constrained quadratic programming, On the convergence rate of scaled gradient projection method, Projected gradient algorithms for optimization over order simplices, Reconstruction of sparse signals in impulsive disturbance environments, Duality-based algorithms for total-variation-regularized image restoration, A Reduced-Space Algorithm for Minimizing $\ell_1$-Regularized Convex Functions, On the steplength selection in gradient methods for unconstrained optimization, Combined Newton-gradient method for constrained root-finding in chemical reaction networks, Spectral Properties of Barzilai--Borwein Rules in Solving Singly Linearly Constrained Optimization Problems Subject to Lower and Upper Bounds, On the Preconditioned Delayed Weighted Gradient Method, Iterative regularization algorithms for constrained image deblurring on graphics processors, A new analysis on the Barzilai-Borwein gradient method, The 2-coordinate descent method for solving double-sided simplex constrained minimization problems, IMPROVED PROJECTED GRADIENT ALGORITHMS FOR SINGLY LINEARLY CONSTRAINED QUADRATIC PROGRAMS SUBJECT TO LOWER AND UPPER BOUNDS, An improved gradient projection-based decomposition technique for support vector machines, On the working set selection in gradient projection-based decomposition techniques for support vector machines, On the convergence of a Jacobi-type algorithm for singly linearly-constrained problems subject to simple bounds, Steplength selection in gradient projection methods for box-constrained quadratic programs, A limited memory steepest descent method, Multiplicative Updates for Nonnegative Quadratic Programming, Efficient optimization of support vector machine learning parameters for unbalanced datasets, A Two-Phase Gradient Method for Quadratic Programming Problems with a Single Linear Constraint and Bounds on the Variables, On a gradient-based algorithm for sparse signal reconstruction in the signal/measurements domain, A quasi-Newton method for unconstrained non-smooth problems, GPDT, Some projection methods with the BB step sizes for variational inequalities, A predictor-corrector affine scaling method to train optimized extreme learning machine, Complexity and Approximability of Optimal Resource Allocation and Nash Equilibrium over Networks, Nearest neighbors methods for support vector machines

Uses Software

• LOQO
• UCI-ml
• SVMlight
• Kernel-Machines
• CONV_QP
• LIBSVM
• MINOS
• SVMTorch
• MNIST

Cites Work

• Unnamed Item
• An algorithm for a singly constrained class of quadratic programs subject upper and lower bounds
• A modified projection algorithm for large strictly-convex quadratic programs
• Modified two-point stepsize gradient methods for unconstrained optimization
• Nonmonotone globalization techniques for the Barzilai-Borwein gradient method
• 10.1162/15324430152733142
• R-linear convergence of the Barzilai and Borwein gradient method
• Two-Point Step Size Gradient Methods
• Gradient Method with Retards and Generalizations
• Interior-Point Methods for Massive Support Vector Machines
• Nonmonotone Spectral Projected Gradient Methods on Convex Sets
• A Nonmonotone Line Search Technique for Newton’s Method
• Efficient SVM regression training with SMO
• A simple decomposition method for support vector machines