The local search algorithm explores and evaluates different solutions (search space) by applying local changes until an optimal or secure solution is achieved. Local search algorithms are widely applied to numerous complex computational problems, including problems in computer science (particularly artificial intelligence), mathematics, operations research, engineering, and bioinformatics. Some examples of local search algorithms are WalkSat, the two-option algorithm for the street vendor problem, and the Metropolis-Hastings algorithm. Below, we list several specific heuristics to improve local search, with the main objective of escaping local minimums.
Usually, moving from one state to the next involves only a local change in the value of a single variable, hence the name local search. Local search algorithms move from one solution to another in the candidate solution space (the search space) by applying local changes, until a solution that is considered optimal is found or a deadline has elapsed. Since the local search depends on the initial solution, there is a high chance of being stuck in a local optimal. Improvements to the SLS can be made in the selection of the initial task and in the nature of the local changes considered, or by trying to escape local minimums.
The local search algorithm explores and evaluates different solutions (search space) by applying local changes until an optimal solution is achieved or certain iterations are calculated.