How to build a knowledge system by self learning deep learning?

In recent years, deep learning has become more and more popular, and many remarkable achievements have been made, such as computer vision, natural language processing, and various **.

For beginners who want to learn deep learning on their own, how to quickly establish a knowledge system?

We know that the establishment of a knowledge system requires continuous and in-depth accumulation and sorting on the basis of focusing on structure, and in-depth thinking through structured thinking. Self-taught deep learning, if you want to quickly build a knowledge system, the best way is to read**.

At present, there are many ** in the field of deep learning, both focusing on theory and application. If you are a novice, it is recommended to start with the application type**, because the impact of this type of literature will be more specific and easier to read.

When reading **, you should start with the introduction, followed by the summary. Don't read the abstract directly, as this may bias your opinion about what is really stated in **. When reading the full text, pay close attention to the research methods and conclusions, find ways to make the conclusions and abstracts relevant to your thinking about the **, and know how this ** fits into the relevant research field.

It is necessary to identify the key points of the research background, identify the problems behind the research, and then find out the solutions used in the research.

In addition, discussing content with people who are more experienced in related materials can help you see the subtleties and find those parts that you should focus more on, which will help you better build a deep learning system.

In order to help students master deep learning technology faster and become high-end talents in the field of artificial intelligence, Zhonggong Education and experts from the Institute of Automation of the Chinese Academy of Sciences have launched the artificial intelligence "Deep Learning" course, so that everyone can learn and truly grasp the principles behind machine learning models and algorithms.

Clarify the subject matter and purpose of the body of knowledge. Before building your own knowledge system, you should be clear about the theme and goals of your own system, only by finding the direction of your own system, finding the part you are most willing to learn, and being able to make yourself more professional knowledge, can you have enough motivation to persist, and clarifying the theme and goal of your knowledge system can help you quickly Li Qing context, eliminate invalid information, and do not have to waste valuable time on irrelevant content. Find ways to gain knowledge.

Common ways to acquire knowledge include books, classes, the Internet, and communication in daily life. Collation and classification of knowledge. This is the top priority of knowledge management, but also the most difficult part of knowledge management, according to the logic and hierarchy to divide as detailed categories as possible, can be sorted out in the form of handwritten notes, can also be recorded by the method of network notes, specially set up a notebook "with your knowledge project" to name, and then your knowledge points (**, text, documents, etc.) according to the name, as notes into your knowledge system.

Output and application of knowledge.

Hello. It is recommended that you make a mind map first.

Once you've taken out the main stem of the whole book, it's enough to list the details.

If you want to learn deeply, it is best to summarize the knowledge network of your own letter files, and usually find free materials on the Internet, or go to the library to consult the materials, and the knowledge points learned are usually very scattered. To establish a knowledge system is to connect small and scattered knowledge points, classify them, and then touch them bypass. If you can find a teacher's open class that has a knowledge context, this is the most convenient way.

First of all, start from a small knowledge bridge, make a small mind map, and accumulate it, a book can be made into a large mind map, and eventually it will become your own knowledge system. Secondly, you can make a keyword classification.

If you want to build a learning knowledge system, then you must learn to read the table of contents, organize the table of contents, and write a complete table of contents by yourself.

1. Learn to set goals for yourself (big, small, long, short) first, so that there will be a direction for learning; Then allocate time reasonably, formulate targeted learning tasks, and implement them one by one.

2. You can learn to master the ability of speed reading and memorization, and improve the efficiency of learning and review. Speed reading and memorization is an efficient learning and review method, and its training principle is to activate the potential of "brain and eye" and cultivate the formation of eye-brain direct reflection reading and learning mode. For the practice of speed reading memory, see "Elite Special Whole Brain Speed Reading Memory Training", practice with software, more than an hour a day, a month's time, you can double the reading speed, memory, comprehension, etc., and finally improve the efficiency of learning and review, and achieve good results.

If your reading and learning efficiency is low, you can practice it well.

3. Learn to integrate knowledge points. Classify the information you need to learn and the knowledge you need to master, and make it into a mind map or knowledge point card, which will make your brain and thinking clear and easy to remember, review, and master. At the same time, learn to connect new knowledge with what you have already learned, and constantly integrate and improve your knowledge system.

This promotes understanding and strengthens memory.

I have a classmate who has the learning behavior of a scumbag, but has the grades of a scholar, and he is simply a person at the level of a god, and we mortals can only sigh that they are really good at learning. Saying that's the case, sometimes he will learn from the scriptures, and what he tells us the most is the word logic. When you logically link all the knowledge points together, you have a knowledge system of your own, and you will get twice the result with half the effort.

How do you build your own body of knowledge?

The first is the ability to synthesize generalizations. You have to regard the knowledge point of a discipline as a big knowledge point, and there are countless small branches in it to grasp the discipline as a whole. Even if you don't have the ability, every book has a table of contents, and you can look at what the table of contents is based on what chapters are subdivided, and summarize the whole subject according to this.

Take the history of modern and contemporary Chinese literature as an example, the history of modern and contemporary Chinese literature is subdivided into chapters according to the era, such as the 30s and 40s, and each era is divided into **, drama, prose, and poetry according to literary style. When you build up such an overall image, you will get twice the result with half the effort when you learn again.

The second is logical reasoning ability. When you have a holistic picture of the subject you are studying, you should add logical thinking and reasoning according to the characteristics of the subject when you learn. For example, there is a writer named Zhao Shuli in the history of modern and contemporary literature, he spans the modern and contemporary, because of his life experience, he likes to write about the countryside, and then think of the characteristics of the 40s and after the founding of New China, you will know that in the 40s, he will be more inclined to criticize the ignorance of the countryside, after the founding of New China, he prefers to write about the beautiful ideas for rural construction.

In this way, the basic general direction of his work is known.

It is difficult to learn, the difficulty lies in only knowing how to learn, it is not difficult to learn, and it is not difficult to think. Just like the saying of Confucius, "If you learn without thinking, you will be reckless, and if you think without learning, you will die." ”