ML Lecture 0-2: Why we need to learn machine learning?
在Github上也po了这个系列学习笔记(MachineLearningCourseNote),觉得写的不错的小伙伴欢迎来给项目点个赞哦~~
ML Lecture 0-2: Why we need to learn machine learning?
Why we need to learn ML

Many people think: Wow!!! AI is so powerful right now! You see AlphaGO? AI is going to replace human beings as new workforce in various fields, isn’t it? It’s unacceptable!

But in fact, we don’t have to worry about that now, because a new industry will come into being and require for a great amount of people, that is: AI Trainer!

But why do we need AI trainer? Machines can learn by themselves, aren’t they?

This question is just like asking: Why do we need Pokemon Trainer?

Because we remember when watching Pokemon, Pokemon trainers only give orders in the back of Pokemon, and they don’t get involved in fight in flesh!!!

But we know that Pokemon trainers are really important actually,

For example, Pokemon trainers need to choose Pokemon with right properties to fight:

Or it may turn out like this:



Also, AI Trainer need to choose model and loss function for machines. Different model and loss function fit with different problems.


Besides, we know that Pokemon may be naughty and doesn’t listen to Pokemon trainer, like this:




And some models are really difficult to optimize such as deep learning models:

Therefore we would need to council AI expert:

So to train a excellent AI, an excellent AI trainer is necessary!

Let’s fight for the goal of being a excellent AI trainer!!!

ML Lecture 0-2: Why we need to learn machine learning?的更多相关文章
- 李宏毅老师机器学习课程笔记_ML Lecture 0-2: Why we need to learn machine learning?
引言: 最近开始学习"机器学习",早就听说祖国宝岛的李宏毅老师的大名,一直没有时间看他的系列课程.今天听了一课,感觉非常棒,通俗易懂,而又能够抓住重点,中间还能加上一些很有趣的例子 ...
- ML-Framework:ML.NET 0.3 带来新组件
ML.NET在今年微软在Build 2018 会议上宣布的机器学习框架现已正式推出0.3版本 https://github.com/dotnet/machinelearning/releases/ta ...
- ML.NET 0.10特性简介
IDataView被单独作为一个类库包 IDataView组件为表格式数据提供了非常高效的处理方式,尤其是用于机器学习和高级分析应用.它被设计为可以高效地处理高维数据和大型数据集.并且也适合处理属于更 ...
- ML.NET 0.9特性简介
ML.NET 0.9已于上周发布,距离上次0.8版本的发布只有一个多月,此次增加的新特性主要包括特征贡献计算,模型可解释性增强,ONNX转换对GPU的支持,Visual Studio ML.NET项目 ...
- ML.NET 0.8特性简介
本周.NET生态圈内的更新源源不断,除了.NET Core 2.2,ASP.NET Core 2.2和Entity Framework Core 2.2之外,ML.NET 0.8也一并登上舞台. 新的 ...
- ML Lecture 0-1: Introduction of Machine Learning
本博客是针对李宏毅教授在Youtube上上传的课程视频<ML Lecture 0-1: Introduction of Machine Learning>的学习笔记.在Github上也po ...
- 课程三(Structuring Machine Learning Projects),第一周(ML strategy(1)) —— 0.Learning Goals
Learning Goals Understand why Machine Learning strategy is important Apply satisficing and optimizin ...
- [Machine Learning & Algorithm]CAML机器学习系列2:深入浅出ML之Entropy-Based家族
声明:本博客整理自博友@zhouyong计算广告与机器学习-技术共享平台,尊重原创,欢迎感兴趣的博友查看原文. 写在前面 记得在<Pattern Recognition And Machine ...
- [Machine Learning & Algorithm]CAML机器学习系列1:深入浅出ML之Regression家族
声明:本博客整理自博友@zhouyong计算广告与机器学习-技术共享平台,尊重原创,欢迎感兴趣的博友查看原文. 符号定义 这里定义<深入浅出ML>系列中涉及到的公式符号,如无特殊说明,符号 ...
随机推荐
- Codeforces Round #612 (Div. 2)C. Garland
第四次写题解,请多指教! http://codeforces.com/contest/1287/problem/C题目链接 题目大意是有一个数字串挂有1-n n个数字,现在上面缺失了一些数字,让你找出 ...
- NLP(二十二)利用ALBERT实现文本二分类
在文章NLP(二十)利用BERT实现文本二分类中,笔者介绍了如何使用BERT来实现文本二分类功能,以判别是否属于出访类事件为例子.但是呢,利用BERT在做模型预测的时候存在预测时间较长的问题.因此 ...
- 量化投资学习笔记37——《Python机器学习应用》课程笔记10
用KNN算法来进行数字识别,还是用sklearn自带的digits数据集. coding:utf-8 KNN算法实现手写识别 from sklearn import neighbors from sk ...
- CSS-水平居中、垂直居中、水平垂直居中
1.水平居中 水平居中可分为行内元素水平居中和块级元素水平居中 1.1 行内元素水平居中 这里行内元素是指文本text.图像img.按钮超链接等,只需给父元素设置text-align:center即可 ...
- disruptor 链路实战 三
一.创建Event类 Trade import java.util.concurrent.atomic.AtomicInteger; public class Trade { private Stri ...
- VUE二 生命周期详解
vue官网对vue生命周期的介绍 Vue实例有一个完整的生命周期,也就是从开始创建.初始化数据.编译模板.挂载Dom.渲染→更新→渲染.销毁等一系列过程,我们称这是Vue的生命周期.通俗说就是Vue实 ...
- Simulink仿真入门到精通(十六) Simulink基于模型设计的工业应用概述
16.1 Simulink用途概述 在基于模型设计广泛应用于汽车电子嵌入式开发的今天,MBD(Model Besed Design)技术也逐步推广到各种嵌入式控制方面.与传统的嵌入式开发相比,BMD以 ...
- QT使用信号量QSemaphore处理大量数据
实现如下:
- python获取群成员信息
#coding: utf-8 import itchat,datetime from itchat.content import TEXT itchat.auto_login(hotReload=Tr ...
- JS实现滚动区域触底事件
效果 贴上效果展示: 实现思路 样式方面不多赘述,滚动区域是给固定高度,设置 overflow-y: auto 来实现. 接下来看看js方面的实现,其实也很简单,触发的条件是: 可视高度 + 滚动距离 ...