Elamur Groupantsworld-opchoice-selection
The groupant beside world/src/opchoice.rs in the Mova repository: what the code computes, what it never does, how to go back from its outputs, its invariants, cost and checks.
Facts
Implementation material. A Mova code model describes this code at one source revision: partial models, finite checks and their stated scope. It is not a reviewed general exposition, and its labels are shown as written, never raised.
Components
| component | label | where |
|---|---|---|
| approximant | partial source model | forward |
| ast | partial | ../syntax.json |
| composition | partial source model | composition |
| cost | partial source model | cost |
| descriptant | partial source model | epoch-memory |
| invariant | partial source model | invariant |
| tests | partial source model | checks |
Obernants
The typed relations around the forward model: the behaviour complement (what the code never does), the reverse reconstruction (every input compatible with an output) and specialised views.
| id | type | label | where |
|---|---|---|---|
world-opchoice-selection-behavior-complement | behavior_complement | partial source model | behavior-complement |
world-opchoice-selection-reverse-reconstruction | reverse_reconstruction | partial source model | reverse-reconstruction |
world-opchoice-selection-zero-weight-choice | priority_relation | partial source model | zero-weight-choice |
world-opchoice-selection-score-kernel | reverse_reconstruction | partial source model | score-kernel |
Model text
Weights, score, choose, margin and SURE before book loading.
forward
Score sums each feature occurrence with absent weights zero; choose replaces only on strict improvement. margin finds first top-score value and minimum top-minus-score among other values; empty returns 0 and no different value returns 1000.
invariant
choose None iff candidates empty under valid indices. Nonempty margin is nonnegative if subtraction representable. SURE=8 is historical empirical threshold, not a correctness probability.
cost
Choose O(CFlog W), plus generated features; margin retains O(C) score/value pairs and scans linearly.
composition
Candidate enumeration feeds features and first-max scoring. Learner interns the same feature occurrences, preserves candidate order and updates toward the first highest-scoring gold-valued candidate. Text extraction and lexical environment must be fixed. top uses lexical expression ties unlike choose; parent ranking cannot silently substitute one for the other.
epoch-memory
2026-10-30: i128 oracle exposed unchecked arithmetic overflow. Checked operations now omit unrepresentable candidates, including MIN/-1 division; score and training overflow remain bounded-domain assumptions. Dense learner history and first-tie tests preserve existing behavior.
checks
See ../tests.md and native opchoice::wave8_models tests. Models are source-derived partial relations, with finite evidence rather than general proof. Syntax projection excludes macro expansion, name resolution and inferred types. No executable approximant: exact gates and categorical choices have no justified percentage metric here.
behavior-complement
Fixed candidate-order universe excludes later equal-score winners in choose. Positive margin does not certify gold correctness; that property is outside the relation.
reverse-reconstruction
Winner j imposes strict earlier and non-strict later linear score inequalities. Observed margin constrains only differently valued candidates; same-value score variation and many weight vectors remain possible.
zero-weight-choice
All-zero weights select first candidate. A larger division weight selects 6/2=3; equal addition/division weights restore first addition. Empty slice returns None.
score-kernel
For multiplicity vector m, score s admits all bounded integer w satisfying dot(m,w)=s. Absent feature coordinates are unconstrained; duplicate features change coefficients.
This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/world/src/opchoice.rs.groupant/selection.groupant/