《Machine Learning A-Z》- Part 0 - Welcome to Machine Learning

03.08.2021

暑假期间充分利用时间，买了Udemy上的 Machine Learning A-Z 课程，在此记录一下学习内容;)

Part 0

Introduction

Introduction

Introduced vast applications of Machine Learning. For example,  face recognition, voice recognition, Virtual Reality, Robot Dogs learn how to walk using Reinforcement Learning (One of the Machine Learning methods),  play games using the X-box,  recommend systems in Facebook and Netflix etc.,  used in space to recognize certain areas of the world through maps and explore new territories like Mars.

Distinguish AI, ML, DL

lecturer给出了特别简洁明了的解释：

AI is the science that empowers computers to mimic human intelligence. AI is a board field that contains several subfields like Machine Learning, Robotics, and Computer Vision.

ML is the subset of AI that enables machines to improve at tasks with experiences. Note that there are some Rule-based engines that could be classified as AI but they don't learn from experience so they can't be categried as ML. ML categories: Supervised Learning, Unsupervised Learning, Reinforcement Learning

DL is the specialized field of ML that relies on the training of Deep Artificial Neural Networks using large dataset like images and texts. It works by mathematically mimicking the human brain and connecting multiple 'artificial' neurons in a multilayered fashion. The difference between ML and DL is that DL has the ability to extract features automatically.

ML process: 1) Selecting the model to train;  2) Manually performing feature extraction.

DL process: 1) Selecting the architecture of the network; 2) Features are automatically extracted by feeding in the training data(such as images) along with the target class(label).

About Regression

x is called independant variable and y is called dependant variable

Simple Linear Regression	y = b + m * x
Multiple Linear Regression	y = b + m1 * x1 + m2 * x2 + ...
Polynomial Regression	y = b + m1 * x1^2  + m2 * x2^2 + ...

 

---