一、Course Introduction
1、机器学习就是自动找函式
Regression: output is a scalar.
Classification:yes or no
Binary classification:RNN作业
Multi-class Classification:CNN作业
Generation: 产生有结构的复杂东西(文字 图片等)
Seq2Seq、GAN作业
2、Supervised Learning: Labeled Data
unsuperived learning
3、前沿研究
Explainable AI
Adversarial Attack:对抗攻击
Network Compression
Anomaly Detection:异常检测(能够知道自己不知道)
Transfer Learning(Domain Adversarial Learning)
Meta Learning:learn to learn
Life-long learning:终身学习
二、Regression
1、应用
2、Step