ID 原文 译文
1983 针对 RBM 梯度近似的一种计算方法对动量加速不敏感，以及识别效果不理想等问题，本文提出一种基于修正动量的 RBM 算法。 Focusing on the gradient approximation algo-rithm insensitivity to the momentum acceleration and recognition effectiveness in RBM, we propose the algorithm based onmodified momentum using RBM.
1984 该算法结合 RBM 梯度近似方法，通过修改隐单元偏置参数的更新方式，避免 RBM模型中隐单元取值采用概率值时导致模型识别效果不理想、动量加速有限等问题。 When the rule to update the hidden states adopts the probability value instead of sampling abinary value, this calculation method for the RBM gradient approximation leads to the undesirable recognition performance andlimited momentum acceleration. Therefore, we modify the updating rule of the hidden bias to avoid these problems.
1985 同时，在 RBM 预训练阶段采用快速上升的动量方式，以加速网络收敛; Simultane-ously, we use the rapidly ascending momentum method to improve the learning speed in the RBM pre-training phase.
1986 在微调阶段引入缓慢下降的动量项，以避免陷入局部最优点并提高识别效果。 An im-proved slowly descending momentum method is also used in the fine-tuning stage to accurately find the best point, which is far from becoming trapped in poor local optima and improves the classification effect.
1987 本文算法通过在 MNIST 手写数字体，Extended Yale B 和 CMU-PIE 人脸数据库上的数值实验结果表明，提出的算法能够有效地提高计算效率和提高网络泛化能力。 Through the recognition experiments onMNIST dataset, Extended Yale B and CMU-PIE face dataset, the achieved results show that the proposed algorithm can en-hance the computation efficiency and improve the generalization ability of networks.
1988 该算法不仅对 RBM 的应用领域扩展具有十分积极的实际意义，且为深度学习的应用方法提供一种新的研究思路和借鉴。 The algorithm not only extends the appli-cation fields of RBM, but also provides a new research idea and reference for the application method of deep learning.
1989 参数化协方差矩阵估计(Parametric Covariance Matrix Estimation，PCE)方法利用雷达系统参数估计杂波协方差矩阵(Clutter Covariance Matrix，CCM)，显著提升非均匀环境下空时自适应处理(Space-Time Adaptive Processing，STAP)的性能; The parametric covariance matrix estimation (PCE)method uses the system parameters to estimate theclutter covariance matrix (CCM ). It can greatly improve the performance of space-time adaptive processing (STAP)in nonhomogeneous environment.
1990 但是在系统参数和杂波分布存在误差情况下，性能下降严重。 However, the performance of PCE method is seriously degraded when the system parameter information or clutter distribution is in error.
1991 本文提出一种稳健的基于 PCE 方法的STAP 杂波抑制方法。 This paper presents a robust parametric covariance matrix estimation basedSTAP method.
1992 首先利用稀疏恢复方法与 Radon 变换估计杂波分布， First the clutter distribution is estimated by the sparse recovery (SR)and Radon transform.