ConceptEngineering
Subgradient method
A simple iterative algorithm to minimize a nondifferentiable convex function by stepping in the direction of any subgradient scaled by a step size.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Opens up
Concepts that name this one as a prerequisite.
Heavy ball method2Projected subgradient method2Step size rules2Alternating projections algorithm1Constant step length1Constant step size1Convergence proof for subgradient methods1Decomposition methods1Dual decomposition1Dual decomposition for network flow1Dual decomposition rate control1Ellipsoid method interpretation1
Lecturers
1Stephen Boyd
Lectures that assume it
4Connections
17The lectures that define it or assume it, and the concepts next to it.
Concepts9
Show 3 moreShow fewer
Lectures8
- Subgradient Methods: Step-Size Rules, Convergence & Polyak Steps
- Convex Optimization II: Localization and Cutting-Plane Methods
- Convex Optimization II: Analytic Center Cutting-Plane and Ellipsoid Methods
- Subgradients II: Directional Derivatives, Cutting Planes & Quasiconvexity
- Convex Optimization II: Primal and Dual Decomposition Methods
- Subgradient Methods for Constrained Problems