Prompt engineering frames an optimization task in natural language so a large language model can generate improved solutions or instructions. Optimization by prompting (OPRO) applies this idea by using LLMs as optimizers. section 11.5
Representative optimization problems include:
- Maximum coverage, MaxCut, and minimum vertex cover. section 11.5
- Boolean satisfiability, traveling salesperson, and vehicle routing. section 11.5
Figure 11.14 Machine learning (ML) for combinatorial optimization (CO) problems
Visualizes machine learning approaches to combinatorial optimization problems. section 11.5
