PACE：面向自动算法设计的基元感知代码进化
PACE: Primitive-Aware Code Evolution for Automated Algorithm Design
- Southern University of Science and Technology（南方科技大学）
- Shenzhen University（深圳大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对现有自动算法设计中局部逻辑与程序绑定导致的组件评估困难问题，提出PACE方法，通过EAP解耦逻辑，经实验验证可保留算法组件并发现竞争力算法。
AI中文摘要：
基于大语言模型（LLM）的自动算法设计通常将算法作为完整、不可分割的程序进行进化。虽然这种全程序视角简化了搜索空间，但它从根本上将有用的局部逻辑与其宿主程序绑定在一起。因此，当整体程序被丢弃时，有价值的代码片段也会消失，这使得评估单个算法组件的贡献变得极其困难。为解决这一问题，我们提出了基元感知代码进化（PACE），它通过将局部逻辑表示为称为可执行算法基元（EAP）的持久单元，将局部逻辑与完整程序解耦。为实现代码级迁移，PACE 维护一个动态的 EAP 集合。算法进化由基元感知算子驱动，这些算子从结构上保证这些组件的保留和跨程序迁移。为有效评估这些算子，PACE 利用基于父代相对性能提升的汤普森采样，从集合中指导基元选择，无需额外的评估数据集。在四个任务上的实验表明，PACE 能够有效发现有竞争力的算法，同时从结构上保留有价值的算法组件。
英文摘要：
Large Language Model (LLM)-based automated algorithm design typically evolves algorithms as complete, indivisible programs. While this whole-program perspective simplifies the search space, it fundamentally couples the useful local logic to its host program. Consequently, valuable code snippets vanish when the overall program is discarded, making it highly difficult to assess the contribution of individual algorithmic components.To address this, we propose Primitive-Aware Code Evolution (PACE), which decouples local logic from complete programs by representing it as persistent units called Executable Algorithmic Primitives (EAPs). To enable code-level transfer, PACE maintains a dynamic set of EAPs. Algorithm evolution is driven by primitive-aware operators that structurally guarantee the retention and cross-program transfer of these components. To evaluate them effectively, PACE leverages Thompson sampling based on parent-relative performance improvements, guiding primitive selection from the set without requiring extra evaluation datasets. Experiments on four tasks demonstrate that PACE effectively discovers competitive algorithms while structurally preserving valuable algorithmic components.
