.png){kind=icon}
.png){kind=icon}
Return
GPU-accelerated Differential Evolution with gradient polishing and decorrelation-based convergence control
.png){kind=icon}
PRE
AI
A
A
DOI：10.1016/j.swevo.2026.102488.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Hybrid GPU-accelerated differential evolution with gradient-based polishing. • Stiffness-aware batched ODE solver with adaptive partitioning (BGSODE). • Statistical decorrelation-based stopping criterion to prevent redundant local searches. • Fully tensorized pipeline achieves up to a 98.2% reduction in optimization runtime.
Journal
IF:
8.5
Papers:
2.1K
Citations:
1.0W
Organization
Cited Papers