Accelerated Genetic Programming Hyper-Heuristics for Simulation-Based Scheduling via Agentic AI
- Source
- arxiv.org
- Author
- Heyang Thomas Li, Alexander Pletzer, Yuan Tian, Yi Mei, Mengjie Zhang
- Date

Context
AI-assisted refactoring substantially shortened a tested project-scheduling simulation workload while preserving its outputs. The paper abstract reports a runtime reduction from 1,298 seconds to under 200 seconds using Anthropic's Claude, with researchers retaining final control.
The workflow paired bottleneck identification and targeted code changes with representative benchmarks and correctness checks. The authors describe custom Python simulations whose sequential updates, loops and conditional logic limited the benefit of numerical or GPU libraries. The result supports optimizing that implementation through measured, checked changes; unchanged outputs on the tests establish agreement for the tested workload.
articleAutomating ML for Science: Can Frontier Agents Climb Scientific Hills in the Wild?Ming Zhong, Stacy Li, Nicholas Carlini, Matthew Jagielski
articleHumanize: Judgement Engineering for Agentic CodingSihao Liu, Ligeng Zhu, Zijian Zhang, Dongyun Zou, Zhengyang Zhang, Changye Li, Song Bian, Song Han, Tony Nowatzki
articleAuditable Human-Agent Teams for Autonomous X-ray ExperimentationGunik Luthra, Ronald J. Pandolfi, Benjamin T. Diroll, Ricky Ronquillo, Ruipeng Li, Christopher B. Murray, Marcus M. Noack, Kevin G. Yager, Esther H. R. Tsai
Checking sign-in…
Loading comments…