Skip to main content Accessibility help
×
  1. Home
  2. Browse subjects
  3. Computer Science
  4. Pattern Recognition and Machine Learning
  5. Content listing

Refine search

Refine search

Access:
Content type:
Publication date:
Apply   From and To filters
Subject: Show more
Journals Show more
Publishers: Show more
Societies Show more
Series: Show more
Collections: Show more

Actions for selected content:

Select all | Deselect all
  • View selected items
  • Export citations
  • Download PDF (zip)
  • Save to Kindle
  • Save to Dropbox
  • Save to Google Drive

Save Search

You can save your searches here and later view and run them again in "My saved searches".

Please provide a title, maximum of 40 characters.
Cancel
×

2327 results in Pattern Recognition and Machine Learning

Page 40 of 117

  • 3 - Descent Methods

  • Stephen J. Wright, University of Wisconsin, Madison, Benjamin Recht, University of California, Berkeley
  • Book:
    Optimization for Data Analysis
    Published online:
    31 March 2022
    Print publication:
    21 April 2022, pp 26-54

  • Summary

    In this section, we discuss fundamental methods, mostly based on gradient information, that yield descent, that is, the function value decreases at each iteration. We start with the most basic method, the steepest-descent method, analyzing its convergence under different convexity/nonconvexity assumptions on the objective function. We then discuss more general descent methods, based on descent directions other than the negative gradient, showing conditions on the search direction and the steplength that allow convergence results to be proved. We also discuss a method that also makes use of Hessian information, showing that it can find a point satisfying approximate second-order optimality conditions and finding an upper bound on the number of iterations required to do so. We then discuss mirror descent, a class of gradient methods based on more general distance metrics that are particularly useful in optimizing over the unit simplex – a problem that arises often in data science. We conclude by discussing the PL condition, a generalization of the strong convexity condition that allows linear convergence rates to be proved.

Optimization for Data Analysis
  • Stephen J. Wright, Benjamin Recht
  • Published online:
    31 March 2022
    Print publication:
    21 April 2022
  • Optimization techniques are at the core of data science, including data analysis and machine learning. An understanding of basic optimization techniques and their fundamental properties provides important grounding for students, researchers, and practitioners in these areas. This text covers the fundamentals of optimization algorithms in a compact, self-contained way, focusing on the techniques most relevant to data science. An introductory chapter demonstrates that many standard problems in data science can be formulated as optimization problems. Next, many fundamental methods in optimization are described and analyzed, including: gradient and accelerated gradient methods for unconstrained optimization of smooth (especially convex) functions; the stochastic gradient method, a workhorse algorithm in machine learning; the coordinate descent approach; several key algorithms for constrained optimization problems; algorithms for minimizing nonsmooth functions arising in data science; foundations of the analysis of nonsmooth functions and optimization duality; and the back-propagation approach, relevant to neural networks.

Page 40 of 117