1) dominated penalty
劣解罚项
1.
This paper presents a constrained multiobjective optimization method by integrating a Pareto genetic algorithm with dominated penalty concept.
该方法提出了劣解罚项的概念 ,把劣解罚项和约束罚项之和作为设计点的总的罚项并将其转化为适应度 ,采用遗传算法进行求解。
2) rewarding good and punishing bad
奖优罚劣
1.
The evaluation indexes of municipal sewage treatment technology were quantified,and the decision matrix was standardized in the range of complied with the principle of rewarding good and punishing bad.
通过对城市污水处理工艺评价指标的定量化处理,并根据奖优罚劣的原则对初始指标集进行[-1,1]区间线性规范化处理,确定理想最优方案(A+)和理想最劣方案(A-),以此计算出各待选方案与它们的相关系数(ζ)。
4) be punished for one's misdeed
因劣迹而受罚
5) penalty term
惩罚项
1.
The mean squ are error relative to the original pixel gray-values is calculated,at the same time,a penalty term is added into this crite.
相对于原始像素灰度值计算均方误差,同时在该准则中加入了一个反映灰度值变化符号的惩罚项。
2.
In this paper,we present and discuss an online gradient method with a penalty term for three- layer BP neural networks.
本文对三层BP神经网络中带有惩罚项的在线梯度学习算法的收敛性问题进行了研究,在网络训练每一轮开始执行之前,对训练样本随机进行重排,以使网络学习更容易跳出局部极小,文中给出了误差函数的单调性定理以及该算法的弱收敛和强收敛性定理。
3.
The cases for two-layer with a penalty term and three-layer without penalty term have been studied in [1] and [2], respectively.
而加入惩罚项则可提高整个网络的泛化能力,也就是提高测试样本的实验正确率。
6) Penalty
[英]['penəlti] [美]['pɛnḷtɪ]
惩罚项
1.
Convergence of BP Algorithms with Penalty for FNN Training;
带惩罚项的BP神经网络训练算法的收敛性
补充资料：驳驳劣劣
1.莽戆。
说明：补充资料仅用于学习参考，请勿用于其它任何用途。
参考词条