求解器推荐器

推荐器从基准条件、声明的能力、规模、目标类型、预算假设与证据层对求解器进行排名。设置下方的六个选择轴；排名在浏览器中根据构建时内联的求解器元数据计算，并解释每一项契合。它从不声称某个求解器在全局上最佳。

Recommendations rank solvers from declared capabilities and benchmark characterization; the recommender never asserts a globally best solver.

Benchmark familyScheduling family of the target benchmark.ConstraintsConstraint posture the recommendation must respect.ObjectiveOptimization objective the solver must support.ScaleInstance scale band. (context input; not scored at the metadata tier)DeadlineDeadline budget posture. (context input; not scored at the metadata tier)Quality budgetTime-versus-quality budget the operator accepts.Evidence confidenceMinimum evidence confidence the recommendation reports. (context input; not scored at the metadata tier)

8 solvers ranked by declared fit; never a globally best claim.

adpso · metaheuristic family · source: metadata · confidence: metadata-only
Fit: matches the current selection on declared capabilities.
Caveat: success-rate feedback is noisy on small swarms
age-moea-ii · metaheuristic family · source: metadata · confidence: metadata-only
Fit: matches the current selection on declared capabilities.
Caveat: geometry estimate is noisy on tiny first fronts
ant-colony · metaheuristic family · source: metadata · confidence: metadata-only
Fit: matches the current selection on declared capabilities.
apparent-tardiness-cost · dispatching family · source: metadata · confidence: metadata-only
Fit: matches the current selection on declared capabilities.
arithmetic-optimization · metaheuristic family · source: metadata · confidence: metadata-only
Fit: matches the current selection on declared capabilities.
Caveat: every member is regenerated around the best vector each iteration, so the returned order is the best observed rather than a monotone incumbent
artificial-bee-colony · metaheuristic family · source: metadata · confidence: metadata-only
Fit: matches the current selection on declared capabilities.
Caveat: the per-component neighbor move explores slowly on long priority vectors, so many iterations may be needed at large task counts
artificial-fish-swarm · metaheuristic family · source: metadata · confidence: metadata-only
Fit: matches the current selection on declared capabilities.
Caveat: each behaviour re-scores candidate positions, so an iteration costs several schedule evaluations per member
beam-search · constructive family · source: metadata · confidence: metadata-only
Fit: matches the current selection on declared capabilities.

Recommendation source: metadata, learned-model. Confidence is capped at metadata-only — a metadata recommendation is never reported as a learned-model one, and no solver is asserted as globally best.

置信度标签

标签	含义
metadata-only	推荐仅基于声明的能力。
smoke-backed	一次小型确定性运行支撑该接口路径。
pilot-backed	试点活动证据已通过评审。
evidence-grade	完整活动证据已通过统计与证据层关卡。

当前的推荐器包为 metadata-only。

来源	依据	它可携带的置信度
metadata	仅依据声明的求解器能力与基准特征化；无学习模型。	metadata-only
learned-model	在过去最佳求解器结果的有标注语料上训练的监督模型。	在无泄漏的留出评估将其提升至 pilot-backed 或 evidence-grade 之前为 metadata-only

选择特征

选择器针对一组固定的基准特征化特征进行排名。每个特征命名其读取的特征化指标：

特征	来源指标
difficulty	聚合难度评分（争用、过载、依赖深度、求解器敏感性）
heterogeneity	资源类型异质性
objective_conflict	目标冲突指标
uncertainty	不确定性指标
dynamism	释放跨度动态性指标
solver_sensitivity	求解器敏感性指标

无泄漏交叉验证

监督选择器报告留出泛化，绝不报告训练拟合。有标注语料被划分，使得任何实例、任何基准族、任何特征化记录都不会同时出现在训练分区与测试分区中；选择器在其余族上训练，并在留出族上评分。有标注语料源自比较性活动，因此在这些活动提供标注之前，选择器显式报告该延迟，而非声称未经验证的性能。

学习与混合接口

推荐器提供七个具名的学习与混合接口。两个在此实现——监督算法选择与仅基准基线——另外五个（代理辅助搜索、强化学习钩子、超启发式、策略引导修复，以及学习初始化）注册为具名接口，其实现取决于产生训练数据的活动。重型后端为可选 extra；当某个 extra 缺失时，接口回退到其确定性默认值，而非默认导入任何东西。接口目录、其证据策略，以及推荐来源分类法，包含在求解器能力包中。