How to learn artificial intelligence with poor mathematical foundation? __ AI

First, the preface

Nowadays, artificial intelligence has become one of the hottest topics. More and more people are starting to want to learn artificial intelligence, so how to get into AI for students who are not very good at Math Foundation. This article to share the mathematical foundation is not how to learn the basics of artificial intelligence, hope to be able to or already in artificial intelligence on the road you bring help, less take some detours.

second, how to learn artificial intelligence

AI is very broad, contains a lot of direction; Before you learn about AI, you should know what the direction of AI is, what you can do roughly, and then choose a direction you would like to study, which will be much more effective.

Here is a brief introduction to the direction of AI. 1. What data types can AI handle Numeric type data

Numeric data is when you're doing an AI project, in the face of data that needs to be processed is numeric or easily converted to numeric types (such as the Gender field, City field, educational education field of this discrete variable) when we are often called numerical data; The common numerical data are financial transaction data, medical data, loan data, etc. text-type data

Text-type data usually refers to the text field (variable) to extract the meaning of the data is called Text data processing, the common text data have news data, comment data, bar data and so on. Picture type Data

Picture-type data refers to the image in the extraction of the meaning of the picture, such as in the picture to identify the license plate number, the image of the identification of cats and dogs and other animals. Audio Type Data

Audio data refers to the identification of content in audio. video-type data

Video-type data refers to the recognition of feature content by artificial intelligence algorithm in video files.

Recommendation: If you want to learn about AI advice you start with numeric data, because numerical data processing is relatively simple. 2. What are the technical directions in the field of artificial intelligence

The field of artificial intelligence is also divided into many technical directions, and then summarize the common technology in the field of artificial intelligence machine learning

Machine learning (Machine Learning, ML) is a multidisciplinary interdisciplinary, involving a number of disciplines such as probability theory, statistics, approximation theory, convex analysis and algorithmic complexity theory. Specialized in studying how computers simulate or implement human learning behavior in order to acquire new knowledge or skills, and to rearrange existing knowledge structures to continuously improve their performance.
It is the core of artificial intelligence, is to make the computer has the basic way of intelligence, its application in all fields of artificial intelligence, it mainly uses induction, synthesis rather than deduction. Neural Network

Neural network, also known as artificial neural network, is a kind of machine learning, and it is a mathematical model of information processing using a structure similar to the synaptic connection of the cranial nerve. In engineering and academia are often referred to as "neural network" or neural network.
The computational model of artificial Neural networks (Ann) is inspired by the animal's central nervous system (especially the brain) and is used to estimate or can rely on a large number of inputs and general unknown approximation functions. Artificial neural networks are usually presented as interconnected "neurons", which can be computed from input values, and can be machine-learned and pattern-recognized due to their adaptive nature of the system. Enhance Learning

Reinforcement learning is a kind of machine learning. Reinforcement learning, which is based on the theory of animal learning, stochastic approximation and optimal control, is a kind of online learning technology, from environment State to action mapping, which makes agent adopt optimal strategy according to maximum reward value. Agent-aware state information in the environment, Search strategy (which policy can produce the most effective learning) Select the best action, thus causing the change of state and getting a deferred return value, updating the evaluation function, completing a learning process, entering the next round of learning training, repeating cycle iterations until the whole learning conditions are met and the learning is terminated. Natural Language Processing

Natural language processing is an important direction in the field of computer science and artificial intelligence. It studies various theories and methods that can realize effective communication between human and computer using natural language. Natural language processing is a science which integrates linguistics, computer science and mathematics. Therefore, the research in this field will involve natural language, that is, the language that people use everyday, so it is closely related to the study of linguistics, but it has important difference. Natural language processing is not the study of natural language in general, but the development of computer systems which can effectively realize natural language communication, especially the software system. So it's part of computer science.
Natural language Processing (NLP) is a field of computer science, artificial intelligence, and linguistics that focuses on the interplay between computers and human (natural) languages. 3. How to get started learning artificial intelligence

In the previous chapters we learned about the mainstream of AI and how to deal with data types, and then I will share with my own experience how I learn AI. The

first step: You need to master an artificial intelligence domain commonly used programming language, Python or R language can, master one of them; I personally recommend you learn the Python language because Python is hot and powerful;
 Python language can do a lot of things, such as crawler, web development, artificial intelligence, where you only need to spend a week to master the Python Foundation, such as how to define variables, how to manipulate tuples, how to customize functions, etc.

Step two: Determine the direction of learning AI. Artificial intelligence is too broad, you need to choose a small branch of artificial intelligence to learn, and so on to learn better in other technical areas of artificial intelligence to learn; here I recommend you to machine learning in the field of the top ten algorithms commonly used in the machine learning algorithm to try to compute numerical data Because the domain algorithm is practical and relatively easy to get started with.

Step three: Suggest you find a course or a book to learn how to use machine learning algorithms, here I recommend you read a book, "Machine Learning combat", if you just contact the machine to learn to complete all the cases inside, or not enough to understand that it does not matter, is normal, the third step is to let you familiar with machine learning algorithms, As well as knowing the specific code implementation of machine learning algorithms, the following projects are used to deepen learning.

Fourth Step: Master data analysis and processing.
Includes: Missing value analysis, anomaly value analysis, variable correlation analysis, continuous variable discretization analysis, etc.
The reason why we need to master data analysis and processing is because in the machine learning project, most of the time to give your data is not complete, such as the high field loss rate, there is garbage data, data analysis and processing of the role is to help you get a clean and effective data, provide a third step machine learning algorithm to do input parameters.

Fifth step: When you have mastered the data analysis and machine learning algorithms then you can download some project data on the Internet through data analysis and machine learning algorithms to achieve and predict results.

Machine Learning project data suggest you go to the Kaggle website to download project data.
When you have completed all of the above steps with a certain degree of accuracy (such as the accuracy rate of 75% and above), congratulations you have started;
When getting started, there are a lot of things to do, such as model optimization, IT project docking, documentation, etc., I believe that after the introduction of the subsequent content of your self-study problem is not. Summary

The development of artificial intelligence today there are more and more artificial intelligence in the direction of the framework appears, the threshold is also less and less; so want to learn ai don't be afraid to learn, math is not good, first learn to use these frameworks and programming language to complete the designated project, After use, you can then try the selective learning of the higher mathematical formulae commonly used.
and the project to complete an artificial intelligence is divided into a lot of processes, mathematics is only one part of the project; in AI projects if you are not good at math, you can also do other parts of the process, the math can work with the better colleagues in math.

`

Attention to the public number: DT Data Technology blog, more artificial intelligence, large data, web crawler tutorial waiting for you.