Latent Heuristic Search: Continuous Optimization for Automated Algorithm Design

In three linesAutomated heuristic discovery via continuous optimization in latent space. Encoder maps discrete programs to continuous embeddings, differentiable surrogate model predicts performance, invertible normalizing flow regularizes optimization trajectory. Evaluation on TSP, CVRP, KSP, and Online Bin Packing shows competitive results against evolutionary baselines.

Summary generated by Claude — human-verified