ID 原文 译文
45466 许多特征选择算法都存在着选择一些冗余和不相关特征的现象，这是因为它们过分夸大某些特征重要性。 Many existing feature selection algorithm have chosen some redundant and irrelevant features, which is leading to overestimation of some features.
45467 同时，过多的特征会减慢机器学习的速度，并导致分类过渡拟合。 Moreover, more features will significantly slow down the speed of machine learning and lead to classification over-fitting.
45468 因此，提出新的基于前向搜索的非线性特征选择算法， Therefore, a new nonlinear feature selection algorithm based on forward search was proposed.
45469 该算法使用互信息和交互信息的理论，寻找与多分类标签相关的最优子集，并降低计算复杂度。 The algorithm used the theory of mutual information and mutual information to find the optimal subset associated with multi-task labels and reduced the computational complexity.
45470 在 UCI 中 9 个数据集和 4 个不同的分类器对比实验中表明，该算法均优于原始特征集和其他特征选择算法选择出的特征集。 Compared with the experimental results of nine datasets and four different classifiers in UCI, the proposed algorithm is superior to the feature set selected by the original feature set and other feature selection algorithms.
45471 首先，基于三子集传播的积分可分性质，分别构造 ARX 结构分组密码积分的 K 集和 L 集传播方程， Firstly, based on three subsets division property propagation technique, the propagation function of the K-setand L-set of ARX block ciphers was constructed respectively.
45472 其中，经过分组密码轮函数异或操作时，L 集所有向量影响 K 集向量传播； All vectors in L-set affected the propagation of K-set when propagate through xored round key operation.
45473 然后，利用 SAT/SMT 求解器，建立 ARX结构分组密码积分传播方程； With SAT/SMT solver, round reduced integral propagation functions of ARX block ciphers could be established.
45474 最后，遍历满足一定数据复杂度的积分输入，自动化搜索缩减轮数的 ARX 结构分组密码积分区分器。 Finally, by exhausting all possible input integral characteristics with proper data complexity, round reduced integral distinguishers of ARX block ciphers could be found.
45475 利用该方法能高效地自动化搜索 ARX 结构，包括类 SIMON 簇、HIGHT、SPECK 簇和 LEA等分组密码算法的积分区分器。 The proposed method can be used for searching integral distinguishers of ARX block ciphers including SIMON-like family block ciphers, HIGHT,SPECK family block ciphers and LEA effectively.