reproduced from "The Heart of Machines"
machine learning community social media often ask this question:
- How do I start machine learning?
- How do I main fly study?
- What is artificial intelligence? How can I learn it?
- How does artificial intelligence work? Where should I start?
- If I do not have a developer Beijing, how do I start?
- ......
In the face of these problems, YouTube blogger What's AI-Louis Bouchard wrote a complete guide on "How to start machine learning in 2021 with zero foundation", integrating a lot of learning Resources, and most of them are free.
Project address: https://github.com/louisfb01/start-machine-learning
This resource has now received 1.6K stars and is still being updated. Let's take a look at the specific content of this guide.
1. Firstly, Bouchard lists some videos on the field of machine learning and its terminology, and compiled free links, including What's AI's Learn the basics in a minute, Welch Labs' Neural Networks Demystified and 3Blue1Brown Neural networks.
Neural Networks Demystified
of Welch Labs 2. Part II Bouchard further lists some more systematic machine learning introductory courses,Including AI Daniel Wu Enda 's Stanford CS229.
3. Read some excellent online articles. Most of the online articles have been viewed many times. The fact that excellent technical articles stand out indicates that they are recognized by many people. The articles currently listed by Bouchard include:
4. In addition to articles, some books with a complete system can also be read and learned. Bouchard has compiled some online editions of the books:
5. Getting started with machine learning What if there is no relevant mathematical knowledge? This discourages many beginners. In this guide, Bouchard recommends three mathematics courses from Khan Academy : linear algebra, statistics and probability, and multivariate calculus. In addition, he also recommended some books and videos related to mathematics for everyone to learn more structured.
Khan Academy’s linear algebra online course.
6. In addition to the basics of mathematics, some interdisciplinary beginners may lack the basic knowledge of programming. Bouchard mainly organizes some course resources for learning Python for everyone:
7. Like other fields, the courses of famous schools and big cows in the field are very high in gold, such as Turing Award winner Yann LeCun, Wu Enda and others have their own online courses. Bouchard's guide currently organizes the following content
8. After mastering theoretical knowledge, practice is also very important. Bouchard recommended the data modeling and data analysis competition platform Kaggle in the guide. Completing the coding and testing of corresponding topics on Kaggle is a common practice learning method in the machine learning community.
Kaggle platform address: https://www.kaggle.com/
9. In addition, Bouchard has also compiled some community platforms or websites that provide news and information in the field. With the help of these platforms, researchers can access the latest research progress and papers, including reddit , Medium, etc.
Finally, Bouchard sorted out the important issues currently facing the AI field-information about ethics and credible AI, and also summarized some tips for machine learning job interviews in his personal blog: https:/ /www.louisbouchard.ai/learnai/#how-to-find-a-job
https://www.ixiera.com
.
