Revisiting the big valley search space structure in the TSP
نویسندگان
چکیده
The solution space of the travelling salesman problem under 2-opt moves has been characterized as having a big-valley structure, in which the evaluation of a tour is positively correlated to the distance of the tour from the global optimum. We examine the big-valley hypothesis more closely and show that while the big-valley structure does appear in much of the solution space, it breaks down around local optima that have solutions whose evaluation is very close to that of the global optimum; multiple funnels appear around local optima with evaluations close to the global optimum. The appearance of multiple funnels explains why certain iterated local search heuristics can quickly find high-quality solutions, but fail to consistently find the global optimum. We then investigate a novel search operator, which is demonstrated to have the ability to escape funnels at evaluations close to the global optimum. Journal of the Operational Research Society (2011) 62, 305–312. doi:10.1057/jors.2010.116 Published online 1 September 2010
منابع مشابه
Deconstructing the Big Valley Search Space Hypothesis
Abstract. The big valley hypothesis suggests that, in combinatorial optimisation, local optima of good quality are clustered and surround the global optimum. We show here that the idea of a single valley does not always hold. Instead the big valley seems to de-construct into several valleys, also called ‘funnels’ in theoretical chemistry. We use the local optima networks model and propose an ef...
متن کاملSearch Landscape of a Realistic Single-Machine Scheduling Task: Peaks With Big Differences
Some scheduling problems, including the Travelling Salesman Problem (TSP), have a “big valley” search landscape: the best solutions have common building blocks, so crossover works well. Unfortunately, in a realistic single-machine problem (TSP with batching, sequence-dependent setups, and multi-level precedence constraints) the best solutions have little in common. A mutation-only approach
متن کاملLandscapes, operators and heuristic search
Heuristic search methods have been increasingly applied to combinatorial optimization problems. While a speciic problem deenes a unique search space, diierentìandscapes' are created by the diierent heuristic search operators used to search it. In this paper, a simple example will be used to illustrate the fact that the landscape structure changes with the operator; indeed, it often depends even...
متن کاملAPPLICATION OF FINITE ELEMENT MODEL UPDATING FOR DAMAGE ASSESSMENT OF SPACE STRUCTURE USING CHARGED SYSTEM SEARCH ALGORITHM
Damage assessment is one of the crucial topics in the operation of structures. Multiplicities of structural elements and joints are the main challenges about damage assessment of space structure. Vibration-based damage evaluation seems to be effective and useful for application in industrial conditions and the low-cost. A method is presented to detect and assess structural damages from changes ...
متن کاملCharacterizing the Distribution of Low-Makespan Schedules in the Job Shop Scheduling Problem
We characterize the search landscape of the job shop scheduling problem (JSSP), with a focus on schedules whose makespan is optimal or near-optimal. Building on previous work on the ‘big valley’ distribution of local optima, we use special branch and bound algorithms to examine in greater detail the extent to which JSSP search spaces conform to the intuitive picture conveyed by the words ‘big v...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- JORS
دوره 62 شماره
صفحات -
تاریخ انتشار 2011