HSEvo
A heuristic-design loop that adds a harmony-search step and diversity measurements to a genetic algorithm run by an LLM.
The loop
- Proposes
Heuristic code from LLM crossover and mutation, plus harmony-search refinement of numeric parameters
- Judged by
Running the heuristic on benchmark instances for an objective value, plus two diversity metrics
- Steers
Objective and diversity scores steer selection to balance exploration and convergence
Headline result
The authors propose two diversity metrics and use them to compare FunSearch, EoH and ReEvo. They report EoH has higher diversity but unstable scores, ReEvo has good scores but poor diversity, and that HSEvo reaches high diversity and good objective scores while staying cost-effective.
What this does not prove
The abstract gives no numeric gains, and results are on the authors' black-box heuristic-design benchmarks. 'Cost-effective' is the authors' wording.
How it generalizes the harness
A genetic algorithm plus harmony search drives a population of heuristics, and the final heuristic runs without the model.
Classification
| What it proposes | Heuristic |
|---|---|
| Generative model | Pretrained LLM, frozen |
| Algorithmic authority | Benchmark evaluation of heuristic code and diversity metrics |
| Feedback | Score |
| What gets updated | Population of candidates |
| When it runs | Offline, at design time |
| Validity | Empirical (measured) |
| Evidence | Experiments reported |
| Lineage | Derivative |