emcuve_select — EMCUVE — Ensemble MC-UVE

Group: Variable selector · Registry tolerance: 1e-06

Description

EMCUVE ensemble MC-UVE (§18 Phase 5n)

Registry note — R plsVarSel::mcuve_pls called n_ensembles times with deterministic seeds (11 + e) and vote-aggregated. Default _emcuve_select_pls4all path mirrors the same R loop, giving bit-exact mask parity. The C++ splitmix64 EMCUVE kernel is opt-in via legacy=True.

Parameters

Name	Type	Default	Notes
n_components	int	2	current public binding signature
noise_features	int	50	current public binding signature
noise_seed	int	0	current public binding signature
n_ensembles	int	10	current public binding signature
vote_threshold	float	0.5	current public binding signature
n_folds	int	3	current public binding signature

Explanations

Bibliographic source

Han, Q.-J., Wu, H.-L., Cai, C.-B., Xu, L. & Yu, R.-Q. (2008). An ensemble of Monte Carlo uninformative variable elimination for wavelength selection. Analytica Chimica Acta 612(2), 121–125. https://doi.org/10.1016/j.aca.2008.02.032 — extends the MC-UVE procedure of Cai et al. (2008) (stability_select) by aggregating independent MC-UVE rounds through a vote rule.

Mathematical principle

Run multiple independent MC-UVE rounds with different seeds, threshold each independently, then vote across rounds: a feature is selected if it survives thresholding in a majority of rounds. Robust against single-round instability caused by particular bootstrap samples.

The voting rule has a free parameter (majority threshold); the default of \(\lceil R/2 \rceil\) is the median-style majority. Stricter thresholds give smaller but more reliable subsets.

Appropriate uses

Combining Monte Carlo coefficient stability with UVE-style elimination for wavelength screening.

Limits and validation

Noise-variable and resampling choices set the effective cutoff; stochastic stability is not evidence of mechanistic relevance.

Implementation

n4m_feature_selection_emcuve_select.

Sources and provenance

Current implementation: cpp/src/core/emcuve_selection.cpp.

API and bindings

C ABI (ABI 2): n4m_feature_selection_emcuve_select. Use the linked public header for the exact signature, configuration, and result handles.

Python (verified public re-export):

from n4m.feature_selection.wrapper import EMCUVE

Source signature: EMCUVE(*, n_components: int = 2, noise_features: int = 50, noise_seed: int = 0, n_ensembles: int = 10, vote_threshold: float = 0.5, n_folds: int = 3) (n4m/_impl/selection.py).

R (source-verified): emcuve_select(X, Y, n_components, noise_features = NULL, noise_seed = 0L, n_ensembles = 5L, vote_threshold = 0.5).

The source signature has additional required inputs, so no example call is fabricated.

MATLAB / Octave (source-verified): emcuve_select(X, Y, n_components, noise_features, noise_seed, ... n_ensembles, vote_threshold).

The source signature has additional required inputs, so no example call is fabricated.

Registry parity references 📐

• 📐 ref.r_plsvarsel (R · r) — plsVarSel 0.10.0 · strict (rmse_rel ≤ 1e-06) — R plsVarSel::mcuve_pls repeated N times with different seeds, then vote-aggregated. Same algorithm family as pls4all’s EMCUVE. RNGs differ; mask metric ~0=perfect.

Benchmarks

Archived measurement identity. Backend labels in this table are the raw IDs recorded when the benchmark ran (including historical pls4all.* IDs). They preserve measurement provenance and do not describe a current public Python, R, or MATLAB binding; use the source-verified API and bindings section above for current entry points.

Adaptive wall-clock per cell measured against full_matrix.csv. Only backends that implement this method are listed; libraries without the method are omitted.

Verdict  ·  ✓ ref / ≈ ref / ~ shape mark a reference-gate pass at strict / relaxed / qualitative tolerance  ·  ✓ bind = archived binding-harness result agrees with the C++ baseline  ·  ⇄ cross-check = documented by-design selector/RNG/model, noncanonical API/facade convention, or secondary oracle  ·  ✗ divergent  ·  ⚠ error  ·  — not run. The fastest backend per column is marked 🏆.

Reference gate: strict — numeric equivalence (rmse_rel_tol ≤ 1e-06).

Rows tagged with 📐 are the canonical parity references for this method (declared in parity_timing.registry). C++ and external rows show reference parity; archived language-harness rows show binding parity against the C++ backend. Hover the icon for role and tolerance band.

1 thread

Backend	Parity	200×40 (ms)
C++ native · libn4m
pls4all.cpp.blas+omp	✓ J 1.00	1.3 s
Python · archived pls4all benchmark
pls4all.python	✓ J 1.00	1.3 s
pls4all.sklearn	✓ J 1.00	1.98 ms🏆
R · archived pls4all benchmark
pls4all.R	✓ J 1.00	4.51 ms
pls4all.R.formula	✓ J 1.00	5.44 ms
pls4all.R.mdatools	✓ J 1.00	5.25 ms
pls4all.R.pls	✓ J 1.00	5.57 ms
R · external
📐ref.r_plsvarsel	source	922.4 ms

3 threads

Backend	Parity	200×40 (ms)
C++ native · libn4m
pls4all.cpp.blas+omp	✓ J 1.00	1.3 s
Python · archived pls4all benchmark
pls4all.python	✓ J 1.00	1.3 s
pls4all.sklearn	✓ J 1.00	2.00 ms🏆
R · archived pls4all benchmark
pls4all.R	✓ J 1.00	4.69 ms
pls4all.R.formula	✓ J 1.00	5.25 ms
pls4all.R.mdatools	✓ J 1.00	5.99 ms
pls4all.R.pls	✓ J 1.00	6.05 ms
R · external
📐ref.r_plsvarsel	source	918.1 ms

10 threads

Backend	Parity	200×40 (ms)
C++ native · libn4m
pls4all.cpp.blas+omp	✓ J 1.00	1.3 s
Python · archived pls4all benchmark
pls4all.python	✓ J 1.00	1.3 s
pls4all.sklearn	✓ J 1.00	1.90 ms🏆
R · archived pls4all benchmark
pls4all.R	✓ J 1.00	4.22 ms
pls4all.R.formula	✓ J 1.00	5.03 ms
pls4all.R.mdatools	✓ J 1.00	4.95 ms
pls4all.R.pls	✓ J 1.00	5.01 ms
R · external
📐ref.r_plsvarsel	source	907.4 ms

---

See also: benchmark overview · methods index · interactive dashboard