AI For Everyone - Learning Week 1 What is AI?

I have completed the study of AI for Everyone from Andrew Ng.

It is a highly recommended course that layouts the basic idea of Artificial Intelligent from basic terminology to business applications.

I found it very beneficial for people who do not have any computer background and ready want to learn what is AI about.

There are 4 weeks of total study.

Week 1 is focusing on basic industry lingoes such as Machine Learning, Data, AI Company, and some limitations on AI with very interesting examples of what AI can or can't do.

What is AI:
AI is Artificial Narrow Intelligent which only focuses on a specific definite repetitive task rather than Artificial General Intelligent like the terminator in a sci-fi movie.

Machine Learning:
It is a type of AI of input to output mappings such as supervised learning. When there are more data input to machine, the more accurate result you can get from it.

There is also a level of AI that will increase its performance: traditional AI>small neural net>medium neural net>large neural net.

Although I hope Andrew Ng can go deeper on what is traditional AI compares with other types of AI.

Data:
There are two types of data: unstructured and structure. Unstructured data are images, audio, text and structure data that are sorted and listed on the table.

Data needs to be cleaned before handed to AI team.

Machine Learning vs. Data Science:

Machine learning is a "field of study that gives computers that ability to learn without being explicitly programmed".

Data Science is "science of extracting knowledge and insights from data".

Deep learning:
input->neural network->output

Neural network is like a black box with complex mathematics to figure out how to price by feeding data.

AI vs Machine Learning vs Deep Learning vs Data Science

AI Company:
Shopping mall + website ≠ Internet Company
Any company + deep learning ≠ AI Company

Unlike traditional company which CEO makes all important decisions, engineer and produce manager at the front line make all decisions that can impact the company operation directly.

AI Transformation:

1.Execute pilot projects to gain momentum

2. Build an in--house AI team

3. Provide broad AI training

4. Develop an AI strategy

5. Develop internal and external communications

Strength of ML:

1. Learning a "simple" concept

2. Lots of data available 

Weakness of ML:
1. Learning complex concepts from small amounts of data
2. It is asked to perform on new types of data

Neural Network:
It is a mapping with predefine factors input with a large number of possible combinations to determine outputs.

Key Takeaway:

After taking week 1 course, I realize AI is artificial narrow intelligent (ANI) that is created to optimize repetitive work with more accurate and specific purpose rather than media was portrayed. There are still far away to reach Artificial General Intelligent (AGN) or even Artificial Super Intelligent (ASI) which the machine surpasses human beings. 

Data Science is another filed that collects and organizes data for AI team to use in the later project phase. And big data is nothing but a channel that can effectively collect the amount of useful data to feed into machine learning and get the result the team plan to achieve.  

Therefore, AI will not take over our job and they can help our job to achieve a new level of productivity with short amount of effort.