Module argmin::solver::linesearch

Line search methods

Line searches are given a position in parameter space and a direction. They obtain a step length in this direction which fulfills a given set of acceptance conditions.

These methods are often an integral part of other methods, such as gradient descent. Each algorithm which implements the LineSearch trait can be used in these optimization methods.

Available line searches

• Backtracking line search
• More-Thuente line search
• Hager-Zhang line search

References

[0] Jorge Nocedal and Stephen J. Wright (2006). Numerical Optimization. Springer. ISBN 0-387-30303-0.

[1] Jorge J. More and David J. Thuente. “Line search algorithms with guaranteed sufficient decrease.” ACM Trans. Math. Softw. 20, 3 (September 1994), 286-307. DOI: https://doi.org/10.1145/192115.192132

[2] William W. Hager and Hongchao Zhang. “A new conjugate gradient method with guaranteed descent and an efficient line search.” SIAM J. Optim. 16(1), 2006, 170-192. DOI: https://doi.org/10.1137/030601880

Modules

• condition
  Acceptance conditions

Structs

• BacktrackingLineSearch
  Backtracking line search
• HagerZhangLineSearch
  Hager-Zhang line search
• MoreThuenteLineSearch
  More-Thuente line search

Traits

• LineSearch
  Line search trait