AI 中文总结
JANUS是即插即用的黑箱优化填充模块，实时估计局部雅可比引导搜索，提升了CMA-ES等宿主优化器在BBOB、多目标等任务上的性能，无需离线训练。
AI 中文摘要
CMA-ES、DE和多目标进化算法等种群优化器主要通过基于标量或排序的选择信号驱动搜索：此类信号仅表明一个候选解优于另一个，却未说明改进的局部方向。JANUS（雅可比对齐牛顿统一搜索，Jacobian-Aligned Newton-Unified Search）是即插即用的填充模块，可在不替换宿主优化器的情况下提取缺失的局部几何信号。它从近期评估轨迹估计局部雅可比，同一雅可比既生成阻尼高斯-牛顿利用候选解，又生成保持迹的探索度量，将宿主每代的部分候选解槽位用于几何引导填充，而非在宿主预算外额外消耗评估资源。与元黑箱优化（MetaBBO）方法不同，JANUS无需离线训练或任务分配，仅从当前运行中实时估计几何信息，同时宿主完全掌控选择、生存、协方差自适应和步长控制。在相同协议的对比中，JANUS使宿主CMA-ES在d∈{30,100,500}的16个BBOB函数中，11-15个函数的性能得到提升；在完整的NN-BBO/MetaBBO基线对比中，d=500时JANUS在16个函数中的13个上达到最佳平均误差，且无训练成本，在d=1000的BBOB子集上比宿主获得936倍的几何平均提升。在结构化和多目标任务中，JANUS在1135维无人机路径规划任务上取得最佳平均代价（比最强基线低12.8%），并在12/38个多目标任务上提升SMS-EMOA/AGE-MOEA2宿主的性能，且无显著退化。代码可通过该网址获取。
英文摘要
Population optimizers such as CMA-ES, DE, and multi-objective evolutionary algorithms drive search mainly through selection signals that are scalar or rank based: such a signal indicates that one candidate outperforms another, but not the local direction responsible for the improvement. JANUS (\emph{Jacobian-Aligned Newton-Unified Search}) is a plug-and-play infill module that extracts this missing local geometric signal without replacing the host optimizer. It estimates a local Jacobian from the recent evaluation trace; the same Jacobian yields both a damped Gauss--Newton exploitation candidate and a trace-preserving exploration metric, reserving a fraction of the host's per-generation candidate slots for geometry-guided infill rather than spending evaluations on top of the host's budget. Unlike MetaBBO methods, JANUS needs no offline training or task distribution, estimating this geometry on the fly from the current run alone, while the host keeps full control of selection, survival, covariance adaptation, and step-size control. Under same-protocol comparisons, JANUS improves the CMA-ES host on \textbf{11--15/16} BBOB functions across $d\in\{30,100,500\}$. It also attains the best mean error on \textbf{13 of the 16} functions at $d{=}500$ in the complete NN-BBO/MetaBBO baseline comparison, with no training cost, and yields a $936\times$ geometric-mean improvement over the host on a $d{=}1000$ BBOB subset. On structured and multi-objective tasks, JANUS gives the best mean cost on 1135-dimensional UAV path planning ($-12.8\%$ vs.\ the strongest baseline), and it improves SMS-EMOA/AGE-MOEA2 hosts on 12/38 multi-objective tasks with zero significant regressions. Code is available at https://github.com/hongyuanyu/JANUS.
Comments27 pages, BBO method
