Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Latent Heuristic Search: Continuous Optimization for Automated Algorithm Design

About

The integration of Large Language Models (LLMs) into evolutionary frameworks has established a new paradigm for automated heuristic discovery. Despite their promise, these methods typically search in the discrete space of program syntax, relying on stochastic sampling to navigate a highly non-convex optimization landscape. This work proposes a continuous heuristic discovery framework that shifts optimization to a learned latent manifold. We employ an encoder to map discrete programs into continuous embeddings and train a differentiable surrogate model to predict performance, enabling gradient-based search. To regularize the optimization trajectory, an invertible normalizing flow maps these embeddings to a structured Gaussian prior, where we perform gradient ascent. The resulting optimized latent vectors are projected through a learned mapper into soft prompts, which condition a frozen LLM to synthesize novel executable heuristics. We evaluate the proposed method on the Traveling Salesman Problem (TSP), the Capacitated Vehicle Routing Problem (CVRP), the Knapsack Problem (KSP), and Online Bin Packing (OBP). Empirical results demonstrate that continuous latent-space optimization achieves performance competitive with state-of-the-art discrete evolutionary baselines while offering a complementary methodological alternative for automated algorithm design. The implementation code is available at \url{https://github.com/cheikh025/LHS}.

Cheikh Ahmed, Mahdi Mostajabdaveh, Zirui Zhou• 2026

Related benchmarks

TaskDatasetResultRank
Traveling Salesman ProblemTSP-100--
69
Traveling Salesman ProblemTSP-200--
41
Capacitated Vehicle Routing ProblemCVRP-50 (test)
Objective Value13.52
5
Capacitated Vehicle Routing ProblemCVRP-100 (test)
Objective Value23.8
5
Knapsack ProblemKnapsack Weakly Correlated capacity 100 50 items hard instance generation protocols of Pisinger
Objective Value133.3
5
Online Bin PackingOBP-5k
Objective Value2.07e+3
5
Knapsack ProblemKnapsack Strongly Correlated capacity 100 50 items hard instance generation protocols of Pisinger
Objective Value162.1
5
Traveling Salesman ProblemTSP-50
Objective Value6.48
5
Knapsack ProblemKnapsack Uncorrelated, capacity 100, 50 items hard instance generation protocols of Pisinger
Objective Value410
5
Online Bin PackingOBP 1k
Objective Value422.6
5
Showing 10 of 10 rows

Other info

Follow for update