NK Landscape Problems

该数据集包含了不同规模和崎岖度的NK景观问题独立实例，这为评估神经-LS策略与基准局部搜索策略的效果提供了可能。数据集分为三个独立采样的实例集，分别用于训练、验证和测试，这些实例均来源于NK(N, K)分布。该数据集涵盖了12种不同场景，其规模（N取值为{32, 64, 128}）和崎岖度（K取值为{1, 2, 4, 8}）各异。任务旨在评估神经-LS策略在NK景观问题上的性能。

This dataset comprises independent instances of NK landscape problems with diverse scales and ruggedness, facilitating the performance evaluation of neural-LS strategies against baseline local search strategies. The dataset is partitioned into three independently sampled instance sets for training, validation, and testing, respectively, and all instances are drawn from the NK(N, K) distribution. This dataset covers 12 distinct scenarios, featuring varying scales (N takes values of {32, 64, 128}) and ruggedness levels (K takes values of {1, 2, 4, 8}). The core task of this dataset is to evaluate the performance of neural-LS strategies on NK landscape problems.