Dont Be Strict in Local Search!

Local Search is one of the fundamental approaches to combinatorial optimization and it is used throughout AI. Several local search algorithms are based on searching the k-exchange neighborhood. This is the set of solutions that can be obtained from the current solution by exchanging at most k elements. As a rule of thumb, the larger k is, the better are the chances of finding an improved solution. However, for inputs of size n, a na"ive brute-force search of the k-exchange neighborhood requires n to the power of O(k) time, which is not practical even for very small values of k. Fellows et al. (IJCAI 2009) studied whether this brute-force search is avoidable and gave positive and negative answers for several combinatorial problems. They used the notion of local search in a strict sense. That is, an improved solution needs to be found in the k-exchange neighborhood even if a global optimum can be found efficiently. In this paper we consider a natural relaxation of local search, called permissive local search (Marx and Schlotter, IWPEC 2009) and investigate whether it enhances the domain of tractable inputs. We exemplify this approach on a fundamental combinatorial problem, Vertex Cover. More precisely, we show that for a class of inputs, finding an optimum is hard, strict local search is hard, but permissive local search is tractable. We carry out this investigation in the framework of parameterized complexity.

💡 Research Summary

Local search is a cornerstone technique in combinatorial optimization, widely employed in artificial intelligence for improving incumbent solutions by exploring a defined neighbourhood. The most common neighbourhood model is the k‑exchange neighbourhood, which consists of all solutions obtainable from the current one by swapping at most k elements. Intuitively, a larger k increases the chance of finding an improving move, but a naïve exhaustive search of this neighbourhood costs O(n^k) time for an input of size n, rendering it impractical even for modest k.

Fellows et al. (IJCAI 2009) investigated whether this brute‑force step can be avoided. They introduced the notion of strict local search: an algorithm must either find an improving solution inside the k‑exchange neighbourhood or declare that none exists, even if a global optimum can be computed efficiently by other means. Their results showed a mixed picture: for several classic problems (including Vertex Cover) strict local search is W