搬运工-AI知识介绍系列

搬运自youtube的视频，多名资深从业者详细介绍了AI基础知识和各个领域的发展。

https://v.youku.com/v_show/id_XNTEyMjU4MDYwOA==.html

Hi! My name is Alejandro Carrillo, and i'm a robotics engineer at an agricultural company.

Specifically my team uses machine learning and robotics and computer vision, to identify the difference between the crops that we eat, and weeds that take nutrients away. We're able
to remove those weeds without any chemicals.

My name is Kate Park and i work at Tesla Autopilot. I build self-driving cars.

Any place where there can be resources used more efficiently, is a place where technology can play a role. But of course, one of the best impactful ways of AI is through self-driving cars.

Have you ever wondered how a computer can recognize a face, or drive a car? Or maybe you've wondered why it's so hard for a computer to tell the difference between a dog and a bagel? Well, it all has to do with something called computer vision: the way machines interpret images.

Let's take a look at a simple example of how computers learn to see. Here are two shapes: an X and an O. At some point you've learned the names for these shapes, but a computer looking at these images for the first time just sees a bunch of little squares, called pixels.

Each pixel has a numerical value for a computer to see. It needs to make sense of these numbers to figure out what is in the picture. In traditional programming, you could tell the computer to check which pixels are filled to decide what shape it sees.

If the center and corner pixels are full, then it's an X.

If the center and corner pixels are empty, then it's an O. Traditional programming works great for this kind of thing, but what about asking the computer to recognize these images? What might the computer think these are? We gave the computer a strict definition of what an X looks like, but these images don't fill all the necessary pixels to fit the definition.

So if the computer doesn't think these are X's at all, in fact the computer thinks these are O's because the corners and center pixels are blank, and that fits the definition of an O that we gave it.

In this example, traditional programming only works some of the time, but with machine learning, we can teach the computer how to recognize shapes no matter their size, symmetry, or rotation.Teaching a computer requires thousands or even millions of examples of training data, and a whole lot of trial and error. So let's start training!

Here are some simple shapes we can use to train the computer to see. At first the computer is completely clueless, and makes a totally random guess from a preset group of options, and it
guesses wrong. But that's okay, because this is where the computer learns. After it makes a guess, the computer is shown the correct answer.

It's like learning with flashcards: sometimes you have to get it wrong before you get it right.

With every guess, the computer looks at each pixel and the surrounding pixels.

It tries to recognize patterns and make rules to help it guess, like if it sees a row of orange pixels next to a row of white pixels, there's an edge. If the computer sees two edges oriented a certain way, say a 90 degree angle, then it's likely to guess that it's looking at a square. It won't get it right every time, but with more trial and error, it will slowly build a more confident guessing algorithm.

Whether it's trying to guess shapes, animals, or any other category, machine learning finds patterns by learning from its mistakes. The training data is used to make a statistical model, which is just a fancy way of saying a guessing machine.

When we give it training data, the guessing machine is tuned and optimized to recognize the pictures we gave it, with the hope that it will then be able to recognize new pictures with the same accuracy. It may seem easy to tell the difference between an X or an O, or to even categorize basic shapes, but most images aren't that simple.

Let's take a look at how computer vision can learn to recognize complex images, or scenes like ones in the real world.

Most complex images can be broken down into small simple patterns. For example, an eye is made up of two arcs and some circles inside. A wheel is made
up of concentric circles and some radial lines.

The way a computer recognizes the patterns in all these pixels, is by using a neural network made of many layers. The first layer of neurons takes pixel values as numerical inputs,

to identify edges. The next few layers of neurons take those edges and try to detect simple shapes, until finally the computer puts
it all together to understand.

It can take hundreds of thousands, or even millions of labeled images, to train a computer vision system. But sometimes even that's not enough! Some face recognition systems have trouble

even seeing people of color, because the system was primarily trained with photos of white people.

Sometimes problems in computer vision are silly, like when a computer gets confused trying to tell the difference between these dogs. Oh wait, that's not a dog!

But it does kind of look like a dog. At least this dog.

But as society relies on computer vision for real problems, like detecting diseases and medical imagery, or helping a self-driving car identify
pedestrians, it becomes increasingly important that we all understand how these systems work and what types of problems they're appropriate for.

Computer vision can open up a miraculous world of possibilities, but a machine is ultimately only as good as the data used to train it.

搬运工-AI知识介绍系列相关推荐

颈椎病的知识介绍系列文章
做IT的人长期伏案工作,很容易得颈椎病,我也有一点,整理这些资料,希望对各位网友有帮助. 大家保重身体啊! 颈椎病的知识介绍(1)-病因颈椎病的知识介绍(2)-分类颈椎病的知识介绍(3)-临床表现 ...
UICC 之 USIM 详解全系列——UICC基础知识介绍
本人就职于国际知名终端厂商,负责modem芯片研发. 在5G早期负责终端数据业务层.核心网相关的开发工作,目前牵头6G算力网络技术标准研究. UICC 之 USIM 详解全系列--UICC(TS102 ...
PMBOK项目管理知识体系指南第七版管理模型介绍系列
本文内容结构一.PMBOK7th 管理模型介绍系列-Cynefin框架简介 1.前言 2.Cynefin框架简介 3.如何确定问题所属区域 4.如何处理每个域的问题 5.总结 (本文作者:苏宁金融- ...
MapboxGL系列（一）基础知识介绍
MapboxGL基础知识介绍文章目录 MapboxGL基础知识介绍前言一.Mapbox是什么? 二.Mapbox常用服务工具介绍 1.mapboxgl.js 2.mapbox studio 3. ...
全球AI技术开放日系列之(七)：走进阿里业务中台
主题: 全球AI技术开放日系列 7(杭州站): 走进阿里业务中台 - 探索阿里电商核心链路智能技术时间: 9月1日 12:30-17:00 报名:点击阅读原文,半价早鸟票限时优惠内容: 全球AI技 ...
【杂谈】有三AI知识星球一周年了！为什么公众号+星球才是完整的？
文/编辑 | 言有三今天,有三AI知识星球开设正好一周年,在有了微信公众号生态后,为什么我们还要做一个知识星球呢?为什么我们说每一个核心的有三AI粉丝一定是在知识星球里呢?今天来总结一下我们知识星球 ...
中秋假期快乐！不发技术文章，有三AI知识星球半价三天
之前在朋友圈说好的中秋节三天不发文章,让大家也好好休息充电一下.昨天收到了不少粉丝朋友的中秋节祝福,为了感谢铁粉们,这个假期我们还是送大家一点东西吧. 月饼咱们就不送了,不值多少钱也不好吃.这三天有三 ...
免费公开课 | AI对抗攻防系列专题，今晚7点第一讲
精选6讲针对人脸识别的AI对抗专题课,搭配实战项目演练,完成项目作业即可获得完课奖品近年来,AI安全问题愈加受到行业关注.在今年6月的智源大会上,清华大学计算机系教授.RealAI 首席科学家朱军就 ...
斯坦福大学CS520知识图谱系列课程学习笔记：第二讲如何构建知识图谱
上一讲我们学习了知识图谱的一些基本概念: 斯坦福大学CS520知识图谱系列课程学习笔记:第一讲什么是知识图谱本节课程关于如何构建知识图谱,因为知识图谱的构建是整个知识图谱领域的一个非常核心且基础的工 ...

搬运工-AI知识介绍系列

搬运工-AI知识介绍系列相关推荐

最新文章

热门文章