1 Introduction
Last Saturday, I watched the Nuggets' class. This class is different from the previous one, because the field involved this time is artificial intelligence, and the content is also how to teach yourself artificial intelligence (AI). For myself, this topic, the field, is a subject of concern but unfamiliar, different front-end engaged in.
2. About artificial intelligence
Before describing the classroom, let's briefly introduce various aspects of artificial intelligence.
2-1. What is artificial intelligence
Quoting Encyclopedia: Artificial intelligence is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. In an attempt to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence, research in this field includes robotics, language recognition, image recognition, natural language processing, and expert systems.
2-2. Why learn artificial intelligence
Why do you want to learn artificial intelligence? Many people may think about it. There is a shortage of artificial intelligence talents and high salary. It's right to think so, because that's how it is now. But in every industry, there will always be an outlet period, a bonus period, and this will not always exist. And if you are not interested and unwilling to accept the challenge, you will study for a high salary, and you will not be able to go far. So, I don't explain why from this point of view.
The reason why I pay attention to artificial intelligence is because artificial intelligence is a trend with broad prospects, and it will gradually penetrate into all walks of life in the future, affecting all aspects of our lives. So I think artificial intelligence is worth learning.
By the way, what are the application areas of artificial intelligence that I understand now?
| field | concrete application |
|---|---|
| in computer vision | License plate recognition, image recognition, face recognition, face beauty, unmanned vehicles |
| situational intelligence | Navigate the best route, user behavior recommendation |
| voice technology | Speech recognition, precise translation, intelligent assistant (speak a sentence, the computer completes a task) |
| production, manufacturing | Automated workshop, smart agriculture |
3. About the classroom
3-1. Themes
Mastering AI self-learning roadmap in one class
3-2. Understand the goal
How can AI change our lives?
The application of AI in Qiniu Cloud?
What basic knowledge do you need to learn AI?
What knowledge is required to participate in AI-related work?
What is the day-to-day job of an AI-related position?
How to teach yourself AI
related resources
4. AI technology innovation
First of all, the first class is the "AI Technology Innovation" brought by Mr. Peng Yao. The content is divided into the following three parts.
4-1. If AI changes our life
Regarding artificial intelligence changing life, there are three main parts, computer vision, language recognition, and speech recognition. The specific products are not listed here.
Applications in computer vision, such as vehicle recognition for navigation, face recognition, and behavior analysis.
In terms of speech recognition, such as intelligent robots and intelligent speakers.
In terms of language processing, online customer service, personal assistant, only question and answer.
There are many excellent products in these fields, which are not listed here. But it is undeniable that because of the emergence of products in these fields, the basic necessities of life for us have been greatly improved.
4-2. Application of AI in Qiniu Cloud
The teacher introduced that Qiniuyun provides innovative and flexible combination scenarios around massive data, and integrates AI into Qiniuyun's product line. At the same time, afterburner develops the exploration and application of video intelligence and data intelligence. Among them, video intelligence includes content auditing (processing of pornographic, violent and terrorist videos to ensure healthy video content), face recognition (identity verification, intelligent security, etc.), video analysis, etc. In data intelligence, including data analysis and decision-making , understanding machine language and emotions, insight into the future, and more.
In addition, the core innovation system of Qiniuyun's artificial intelligence laboratory is also introduced. Introduces the use of AI to achieve content auditing (processing pornography, violence and terror and other video images to ensure healthy content), City Eye (detecting people, objects, and scenes, realizing identity verification, intelligent security, etc.), media resources intelligence (Face recognition for video images, etc., and manual retrieval needs), innovation plan. It also introduces the technical architecture, operation principles, and some successful cases (Momo, BBK, Meitu Xiuxiu, etc.).
4-3. Daily life of AI engineers of Qiniuyun
First of all, there are 7 types of AI engineers, computer vision algorithm engineer, machine learning platform R&D engineer, big data platform R&D engineer, search engine R&D engineer, system architecture engineer, and business architecture engineer. These positions literally know what they are responsible for, although I don't. The teacher briefly introduced their work content in the class, and the achievements include AVA elastic deep learning platform, LEGO big data rich media knowledge base, AI Video OS and so on.
4-4. Problem Review
Because the typing speed cannot keep up with the speaking speed, and the content is too much, I express my own meaning and worry about the wrong expression (actually because I am lazy), so I will not write the answer. If you need to know the answer, please click the link below to watch video.
1. What are the core technical capabilities of Qiniu Cloud AI?
2. What are the possible foundations and challenges for combining machine learning with social and social research?
3. Many articles on artificial intelligence say that China is in a leading position in the field of human beings, but why is the quota information found in the learning process all from abroad?
4. What is the difference between system architecture and business architecture?
5. Can you briefly describe the process of AI development?
6. What application scenario is this AI Video OS used for? Is it to provide service capabilities externally, or to use it internally?
Finally, because this is just a brief introduction of my notes, if you want to know more, you still have to click the link below to watch:
5. Getting Started with AI for Beginners
Then, the second class was "Introduction to AI Beginners" brought by Mr. Shao Jie. In this class, not all application areas of AI are covered, but two areas with wide application are selected: machine learning and computer vision.
5-1. Course content
Before we start, three suggestions are made:
1. Don't wait until you have mastered all the relevant math knowledge before starting 2. Don't collect too much learning materials 3. Do it, do it, do it
Regarding the first suggestion, the purpose of the teacher is to say that because there is too much knowledge, it is difficult to master it all, which will affect confidence, and a lot of knowledge does not necessarily require special familiarity. I want everyone to enter the learning of artificial intelligence as soon as possible. It is true that not only AI, but also the front-end is like this. When someone evaluates whether to be a front-end or a back-end, it is to evaluate what to learn in the front-end and what to learn in the back-end, so it feels a little too much.
Regarding the second suggestion, the teacher feels that the online materials are complicated, the quality is difficult to guarantee, and it is not systematic, and the time cost is too high to collect too much information.
Regarding the third suggestion, this application is universal. Even if I write an article myself, I highly recommend it, and I have repeatedly mentioned that in addition to reading, it is necessary to write, so as to be impressed. If you just read but don't write, the learning process is easy to get confused.
Machine Learning: Using learning algorithms to generate models from data. Simply put, according to the written program (machine algorithm), a model is generated according to a large amount of data. The lecturer also talked about an example: for example, you often receive spam, and next time you receive it, you will analyze the received spam to determine whether it is spam.
Machine learning: generalization (analyzing new data based on existing data), algorithm preference (different models, problems, applications match different algorithms)
Regarding machine learning, K-nearest neighbors are also used to realize an image recognition. But before learning, you need to learn the relevant mathematics knowledge.
5-2. Problem Review
Still the same, only questions and answers watch the video.
1. Do you recommend starting with deep learning?
2. Phtroch and TensorFlow are two machine learning libraries, which one is better for learning?
3. Hands-on is very important, how should you practice it?
4. What is the appropriate language for AI development?
5. May I ask the teacher, what advice do you have for the traditional software development industry (C language) and Xingxing artificial intelligence industry (machine learning direction)?
6. What do teachers think about online machine learning courses, such as coursera, etc. Please tell me
Video link:
5-3. Learning materials
Learning materials mentioned in the video:
books
Nick's "A Brief History of Artificial Intelligence" (this book needs to recognize the author, because the teacher mentioned that there are two books with the same name in "A Brief History of Artificial Intelligence", and Nick is the one recommended)
Miroslav Kubat "Introduction to Machine Learning"
Zhou Zhihua, "Machine Learning" (Watermelon Books)
Aurelien Geron《 Hands-on Machine Learning with Scikit-learn & Tensorflow 》
Ian Goodfellow等《Deep Learning》（花书）
ML 101
Getting Started With MachineLearning (all in one) by 梁劲 sina.lt/f3W8
Machine learning 101 by Jason Mayes sina.lt/f3W3
Online courses
Machine Learning Crash Course developers.google.com/machine-lea…
Professor Li Hongyi, National Taiwan University speech.ee.ntu.edu.tw/~tlkagk/cou…
Professor Ng Enda mooc.study.163.com/smartSpec/d…
Stanford University cs231n cs231n.stanford.edu/
Stanford cs224n web.stanford.edu/class/cs224…
other
scikit-learn Tutorials scikit-learn.org/stable/tuto…
Machine Learning Glossary developers.google.com/machine-lea…
4. Summary
The personal understanding of this class is almost here. But this class can give everyone a general understanding of AI development, and know what AI can do and how to do it. And a guide for AI self-learning, as to whether it is useful or not, this is a matter of opinion.
Finally, I would like to thank the Nuggets and two lecturers for bringing the course to everyone.