RcppEnsmallen v0.1.16.0.1 Released - New LBestPSO Optimizer and Objective Value Control

software-releases
r
Author

TheCoatlessProfessor

Published

August 29, 2019

The RcppEnsmallen package brings to R a Header-Only C++ Mathematical Optimization Library for Armadillo. In particular, Ensmallen is a templated C++ mathematical optimization library (by the MLPACK team) that provides a simple set of abstractions for writing an objective function to optimize.

Well, CRAN was on vacation when the Ensmallen Development Team released this version. So, there was a slight delay in getting it listed on CRAN. Within the new version, there is a new optimizer LBestPSO that was in the works for awhile now (~February 2019). Moreover, there was an API change related to adding an option to avoid computing the exact objective function value at the end of the optimization routine. There was another gcc diagnostic message that was also addressed related to types. The full notes for the release can be found below.

RcppEnsmallen news file entry for version v0.1.16.0.1 (2019-08-29)

  • Upgraded to ensmallen release 1.16.0 “Loud Alarm Clock” (2019-08-09)
    • Add option to avoid computing exact objective at the end of the optimization (#109).
    • Fix handling of curvature for BigBatchSGD (#118).
    • Reduce runtime of tests (#118).
    • Introduce local-best particle swarm optimization, LBestPSO, for unconstrained optimization problems (#86).
  • Fix return type error in PSO (#123)

GitHub Changelog