AI----Introduction to Artificial Intelligence

[Original link] AI----Introduction to Artificial Intelligence

1. Introduction to artificial intelligence

1.1 Elements of the development of artificial intelligence

• data
• algorithm
• Computing power

1.2 The difference between CPU and GPU

• CPU is mainly suitable for I/O intensive tasks
• GPUs are mainly suitable for computationally intensive tasks

1.3 The connection between artificial intelligence, machine learning and deep learning

• Machine learning is the way to achieve artificial intelligence
• Deep learning is a method of machine learning
  

1.4 Artificial Intelligence Development Stage

1.5 Main branches of artificial intelligence

• computer vision
• Speech Recognition
• Text mining/classification
• machine translation
• robot

2. Machine Learning

2.1 Definition of machine learning

Machine learning is to automatically analyze the model obtained from the data and use the model to predict unknown data.

2.2 Workflow of machine learning

• retrieve data
• Basic data processing
• feature engineering
• Machine learning (model training)
• Model evaluation
  • The results meet the requirements and the service is launched online.
  • If the requirements are not met, repeat the above steps.

2.3 Basic concepts of data

• Sample: A row of data becomes a sample
• Feature: A column of data is called a feature
• Target value: Some data have target values, and some data do not have target values.

2.4 Data segmentation in machine learning

• The general data set for machine learning is divided into two parts
  • Training data: used for training and building models
  • Test data: used during model testing to evaluate whether the model is effective
• Division ratio
  • Training set: 70%-80%, such as 75%
  • Test set: 20%-30%, such as 25%

2.5 The meaning of feature engineering

Feature engineers use professional background knowledge and skills to process data so that features can play a better role in machine learning algorithms. Feature engineering will directly affect the effectiveness of machine learning

2.6 Contents of feature engineering

• Feature extraction
• Feature preprocessing
• Feature dimensionality reduction

2.7 Classification of machine learning algorithms

• supervised learning
• unsupervised learning
• semi-supervised learning
• reinforcement learning