大工至善|大学至真分享 http://blog.sciencenet.cn/u/lcj2212916

博文

[转载]【读书1】【2017】MATLAB与深度学习——什么是机器学习(2)

已有 428 次阅读 2018-9-10 10:19 |系统分类:科研笔记|文章来源:转载

 

没过多长时间,他们就意识到自己的判断是错误的。

Well, it did not take very long until theyrealized their misjudgment.

 

如果没有明确的规范或规则,你如何能够识别数字?

How were you able to identify the numberswithout a clear specification or a rule?

 

这是一个很难回答的问题,不是吗?

It is hard to answer, isn’t it?

 

但这是为什么呢?

But, why?

 

这是因为我们从来没有学习过这样的规范。

It is because we have never learned such aspecification.

 

从很小的时候起,我们就知道这是0、那是1

From a young age, we have just learned thatthis is 0, and that this is 1.

 

我们只是这样进行区分,当我们面对各种各样的数字时,我们能能够更好地分辨数字。

We just thought that’s what it is andbecame better at distinguishing numbers as we faced a variety of numbers.

 

我说的对吗?

Am I right?

 

那么计算机呢?

What about computers, then?

 

我们为什么不让计算机来完成相似的学习过程呢?

Why don’t we let computers do the samething?

 

就是这样!

That’s it!

 

祝贺你!

Congratulations!

 

你刚刚已经掌握了机器学习的概念。

You have just grasped the concept ofMachine Learning.

 

机器学习已经被用来解决分析模型几乎不可用的问题。

Machine Learning has been created to solvethe problems for which analytical models are hardly available.

 

机器学习的基本思想是,当方程和定律不可能实现时,使用训练数据来实现模型。

The primary idea of Machine Learning is toachieve a model using the training data when equations and laws are notpromising.

 

机器学习面临的挑战Challenges with Machine Learning

 

我们已经知道机器学习是用来从数据中找到(或学习)模型的技术。

We just discovered that Machine Learning isthe technique used to find (or learn) a model from the data.

 

它适用于涉及智能的问题,如图像识别和语音识别,解决物理定律或数学方程不能实现模型的问题。

It is suitable for problems that involveintelligence, such as image recognition and speech recognition, where physicallaws or mathematical equations fail to produce a model.

 

一方面,机器学习使用的方法是使得模型的产生过程得以实现。

On the one hand, the approach that MachineLearning uses is what makes the process work.

 

另一方面也带来了不可避免的问题。

On the other hand, it brings inevitableproblems.

 

本节将介绍机器学习面临的基本问题。

This section provides the fundamental issuesMachine Learning faces.

 

一旦机器学习从训练数据中找到模型,我们将模型应用到实际的真实数据。

Once the Machine Learning process finds themodel from the training data, we apply the model to the actual field data.

 

该过程如图1-4所示。

This process is illustrated in Figure 1-4.

 


1-4 基于现场真实数据的模型应用Applying a model basedon field data

 

——本文译自Phil Kim所著的《Matlab Deep Learning》


更多精彩文章请关注微信号:qrcode_for_gh_60b944f6c215_258.jpg




http://blog.sciencenet.cn/blog-69686-1133906.html

上一篇:[转载]【2018.07】异构无线网络中车辆到基础设施通信的高效切换方法
下一篇:[转载]【源码】块坐标下降算法BCD——让块坐标下降法实现更快

0

该博文允许注册用户评论 请点击登录 评论 (0 个评论)

数据加载中...

Archiver|手机版|科学网 ( 京ICP备14006957 )

GMT+8, 2018-11-18 03:12

Powered by ScienceNet.cn

Copyright © 2007- 中国科学报社

返回顶部