Random Restart — A Re-exploration Strategy for Escaping Local Optima

Random Restart is a meta-strategy that improves search and optimization by running the same algorithm multiple times from different initial states and selecting the best solution found. In complex search spaces, a single run can easily become trapped in an unfavorable local optimum because the result heavily depends on the starting point. Random Restart reduces this dependency by providing multiple opportunities to explore different regions of the solution space and increases the chance of discovering a better solution.

08/02/2026

Simulated Annealing — Search that briefly allows worse moves to escape a local optimum

Simulated Annealing is a local search algorithm designed to reduce the chance of getting stuck in a local optimum by probabilistically accepting some worse moves during the search.

10/28/2025