Adaptive Regularization using Cubics for Optimization
Overview
Implements cubic regularization methods (ARC) for local optimization problems common in statistics and applied research. Provides robust handling of ill-conditioned, nonconvex, and indefinite Hessian problems with automatic saddle point escape. Supports box constraints; linear equality constraints are planned for a future release.
This is the first CRAN release of **arcopt**. The headline feature is
the tri-modal solver shipped in v0.2.0 (cubic / trust-region fallback /
quasi-Newton polish), wrapped with the tiered control surface
documented under ?arcopt (Tier 1) and ?arcopt_advanced_controls
(Tier 2). See the v0.2.0 entry below for the full feature description.
Other changes since v0.2.0
- Cubic solver now reuses its eigendecomposition at end-of-iteration detector checks, removing redundant work and trimming roughly 0.3% of total runtime on large-scale benchmarks.
- Added benchmarking and profiling scaffolding under
benchmarks/for identifying C++ port candidates. Excluded from the package build; available in the GitHub source tree.
From NEWS.md in the release
Install
Health
- OK13 OK2026-09-08
- ERROR12 OK, 1 ERROR2026-09-07
- OK13 OK2026-08-05
- NOTE12 OK, 1 NOTE2026-08-01
- OK7 OK2026-04-30
Documentation
- Help pages
- 30 (28 internal)includes a package overview page
- Examples that run
- 0%1 only when donttest examples are enabled (R CMD check --as-cran)
- Exported functions with examples
- 100%
- Documented parameters
- 78%
- Return-value docs
- 100%
- Help pages citing references
- 13%
Downloads
Repository
- PRs merged (all time)
- 100%
- Median days to close, 50 most recently updated closed issues
- not measured
- Median days to close, 50 most recently updated closed PRs
- not measured
10 development-tooling and community-health practices detected across 5 families in the upstream repository
Checks run against github.com/marcus-waldman/arcopt on 2026-10-18.
The R CMD check workflow runs on Linux, macOS and Windows, and on R-devel
GitHub Pages site last deployed 2026-04-28 · The repository has a site deploy workflow
Show all practicesHide 1 practice
This repository shows signs of AI coding tools.
- Claude Codefirst seen 2026-01-03Seen in the repository: CLAUDE.mdNamed in .Rbuildignore: .claude, CLAUDE.mdNamed in .gitignore: .claudePull request it wrote: opened by a project memberCommits: at least 46 crediting it in the message
- Claude Opus 4.718 commits · Apr 2026
- Claude Opus 4.512 commits · Jan 2026
- Claude Sonnet 4.58 commits · Jan 2026
- Claude Haiku 4.54 commits · Jan 2026
- Claude Sonnet 4.64 commits · Apr 2026
Dependencies
Code & Tests
People & History
- 0.3.0Latest2026-04-29 · current release
- RR 4.6.0 released · 2026-04-24
Package metadata
- First published
- 2026-04-29
- Total releases
- 1 / 1 yrs
- License
- MIT + file LICENSE OSI
- Minimum R
- ≥ 4.1.0
- Download size
- 110 KB
Cite
Cite this package
Run in R for the authors' entry:
citation("arcopt")The package provides a citation file, so citation() returns the authors' own entry.
It points to 10.1007/s10107-009-0286-5.
Cite the R Observatory
For a number measured here: a download total, a coverage figure, an archival date.
From data release v2026-10-21, which the citation names so these numbers can be found later. More on citing and the projects behind them.