一种多任务贝叶斯优化的自适应知识迁移方法
By introducing an explicit knowledge transfer mechanism and a kernel-based autoencoding mechanism, this method solves the problem of complex inter-task relationships in multi-task Bayesian optimization, achieving efficient knowledge sharing and improved optimization accuracy. It is an adaptive knowledge transfer method suitable for multi-task Bayesian optimization.
CN120996138BActive Publication Date: 2026-07-17SOUTH CHINA UNIV OF TECH
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SOUTH CHINA UNIV OF TECH
- Filing Date
- 2025-08-05
- Publication Date
- 2026-07-17
Smart Images
Figure CN120996138B_ABST
Abstract
本发明公开了一种多任务贝叶斯优化的自适应知识迁移方法，包括以下步骤：利用内核化自动编码机制来捕捉数据集之间的非线性关系，以度量任务之间的相似性；通过启发式规则动态调整辅助任务选择的优先级，确保任务间的选择性协作并减少干扰。运用贝叶斯优化建模和优化黑盒函数，实现对任务真实评估次数的减少。本发明引入显式知识迁移机制，运用非线性映射矩阵转化辅助任务的最优解，实现多个相关任务之间高效的知识迁移与协同优化，显著提高了优化效率和资源利用率。
Need to check novelty before this filing date? Find Prior Art