# 论文的语言

371人阅读 评论(0)

## 1. 介绍公式

ΔoutputjoutputwjΔwj+outputbΔb

• is, denote, with respect to(wrt, 关于)

• 和式的介绍：where the sum is over all the weights, wj$w_j$,
• 记号的意义：and outputwj$\frac {\partial \, \mathrm{output}}{\partial w_j}$ and outputb$\frac {\partial \, \mathrm{output}} {\partial b}$denote partial derivatives of the output with respect to wj$w_j$ and b$b$, respectively.
• 阐述：we denote the number of neurons in this hidden layer by n

## 2. 动词篇

• follow && employ：Our design and implementation follows the approaches employed in …
• comprise：In this work, we propose a very simple deep learning network for image classification which comprises only the very basic data processing components:
• emerge：with many articles authored or co-authored by Chinese researchers emerging both from Chinese institutions and from from institutions outside of China.

## 3. 名词篇

• design and implementation && approach：Our design and implementation follows the approaches employed in …

## 4. 副词 & 连词篇

• however：可放在句首，可放在句中做插入语

In our version, however(表示一种对比转折), we used a data structure that is more type-safe.

In addition, we used many C++ language features…

• in addition to：除….之外，还有。与in addition不同，in addition to是一个介词。

In addition to the data members inherited from class CenterCluster, class SubspaceCluster has a protected data member, which is ….

• hence：表达因果关系，因此。。。，与so的不同在于，so后的从句作为一个子句。hence引导的句子可单独作为一个句子。（这种用法很类似，Since和because）

Hence it is natural to design the base cluster class as a collection of records.

## 5. 形容词

• Image classification based on visual content is a very challenging task, largely because there is usually large amount of intra-class variability, arising from

## 5. 常用表达

• 没有关于XX的统一的定义：In cluster analysis, there is no uniform definition for a cluster.
• XX有着不同的形式：Clusters comes with many different forms.

• 仅举这么几例，“不一而足”：For example, there are one-center clusters, multi-center clusters, fuzzy clusters, subspace clusters, to name just a few.

• 也就是说That is, 可放在子句的中间作为插入语，也可置于句首，作为一个独立的句子的开始。
Since class Cluster does not contain any pure virtual functions, it is not an abstract class, That is, we can create objects from class cluster.

• 从….的观点来说：From a technical point of view,
• 在特定条件下：Under certain conditions

this assumption is violated
we can skip the first step and proceed with the calculation of the mean vecs进行，执行

• we relax the condition：These studies assume the vectors of the different channels to be jointly sparse. In this paper, we relax the condition.
• Numerous efforts have been made to counter the intra-class varaiability by manually designing low-level features for classification tasks at hand. Representative examples are

## 6. 标点篇

• 英文状态下的小括号

左括号(之前和右括号)之后要留一个空格，但其后（wrt，主要是对于右括号而言）如果是另外的标点，比如,或者.，则可省略相应的空格。

In data clustering, a cluster is a collection of objects or records that share the same properties(Carmichael et al., 1968; Bock, 1989; Everitt, 1993).（左括号之前有一个空格，右括号之后，因为是一个标点，可以不加空格）

## 7. 段内结构

• it is followed by
• First && Then && Finally

In this chapter, we introduce the design and implementation of clusters. First, we introduce the design of a single cluster. Then, we introduce how we design the result of a partitional clustering algorithm. Finally, we introduce the design of the result of a hierarchical clustering algorithm.

• 插入语的位置：

• The disadvantage, of course, is that this is computationally very expenisve when we need to esitmate multiple parameters
• use of proposal distributions (e.g. rejection and importance sampling).
|

## 8. 实验环境的介绍

• all numerical experiments are performed on the personal computer with a 3.40GHz Intel(R) Core(TM) i3-3240 CPU and 3.15G bytes of memory. This computer runs on Windows 7 with MATLAB 8.1.0.

## 9. 算法的描述

Given an initial guess for θ$\theta$ with positive probability of being drawn, the Metropolis-Hastings algorithm proceeds as follows

• Choose
• Calculate
• If
• Repeat the earlier steps

• 中英文互译时，有时需要改变原文中的句式结构，有时需要转换表达方式。这种结构以及表达方式的转换是鼓励的，当沿着原文的结构或者含义进行翻译出现障碍时，应当尝试转换。

1. we can maximize the likelihood (or, equivalently, minimize the loss) to find θ$\theta$ within a frequentist paradigm.

2. 施加影响：
to exert influence on …
It’s clear on examination that the outliers are exerting a disproportionate influence on the fit.

3. The variety of possible loss functions is quite literally infinite, but one relatively well-motivated option is the Huber loss.

4. A and B are C respectively
Ridge regression and the lasso are regularized versions of least squares regression using 2$\ell_2$ and 1$\ell_1$ penalties respectively, on the coefficient vector.

5. account for：解释
The Bayesian approach to accounting for outliers generally involves modifying the model so that the outliers are accounted for. For this data, it is abundantly clear that a simple straight line is not a good fit to our data. So let’s propose a more complicated model that has the flexibility to account for outliers. One option is to choose a mixture between a signal and a background:

P({xi},{yi},{ei}|θ,{gi},σB)=gi2πe2iexp[(y^(xi|θ)yi)22e2i]+1gi2πσ2Bexp((y^(xi|θ)yi)22σ2B)

What we’ve done is expanded our model with some nuisance parameters: {gi}$\{g_i\}$ is a series of weights which range from 0 to 1 and encode for each point i the degree to which it fits the model. gi=0$g_i=0$ indicates an outlier, in which case a Gaussian of width σB$\sigma_B$ is used in the computation of the likelihood. This σB$\sigma_B$ can also be a nuisance parameter, or its value can be set at a sufficiently high number, say 50.

6. 足够高：
its value can be set at a sufficiently high number, say 50.

there is a tradeoff between query complexity and query stability.

8. 细节保持
detail-preserving，fine detail-preserving

9. Dissimilarity Measure or Similarity Measure

10. by optimizing an objective function that includes two terms: one that measures the signal reconstruction error and another that measures the sparsity.

11. suffer from multiple local minima,

12. The proposed approach outperforms the standard discriminative methods and the standard sparse representation in the case of corrupted signals.

13. The problem solved by the sparse representation is to search for the most compact representation of a signal in terms of linear combination of atoms in an overcomplete dictionary.
（as for，in terms of，as far as concerned）

In terms of quality, A is better than B.
As far as quality is concerned, A is better than B.

14. The rest of this paper is organized as follows

15. The rest of this paper is organized as follows. Section 2 reviews the problem formulation and solution for the standard sparse representation. Section 3 discusses the motivations for proposing SRSC by analyzing the reconstructive methods and discriminative methods for signal classification. The formulation and solution of SRSC are presented in Section 4. Experimental results with synthetic and real data are shown in Section 5 and Section 6 concludes the paper with a summary of the proposed work and discussions.

16. Inspired by this idea.

17. Lets Assume an arbitrary Markov Chain P$P$ with infinite states on [0,1]$[0,1]$ having transition Matrix Q$Q$ such that（满足） Qij=Qji=$Q_{ij} = Q_{ji} =$ All entries in Matrix.

18. One advantage of this is that the prior does not have to be conjugate (although the example below uses the same beta prior for ease of comaprsion), and so we are not restricted in our choice of an approproirate prior distribution.

说的是 P(X)$P(X)$的计算：

p(X)=d(θ)p(θ)p(X|θ)

使用数值积分（numerical integration）的方式（on a grid of values for θ$\theta$）进行计算：
p(X)=θp(θ)p(X|θ)

19. All ocde will be built from the ground up to ilustrate what is involved in fitting an MCMC model, but only toy examples will be shown since the goal is conceptual understanding.

0
0

* 以上用户言论只代表其个人观点，不代表CSDN网站的观点或立场
个人资料
• 访问：2096692次
• 积分：57969
• 等级：
• 排名：第48名
• 原创：3931篇
• 转载：38篇
• 译文：1篇
• 评论：139条
博客专栏
 Spark 入门 文章：5篇 阅读：6120
 趣味数学 文章：15篇 阅读：25737
 Python数据结构与算法 文章：7篇 阅读：8796
 小议设计模式 文章：10篇 阅读：6485
 大话面向对象 文章：10篇 阅读：3816
文章分类
我的微博
阅读排行
最新评论