Yufeng Zhao, Lihao Liu

2022/12/6

This lecture is an introduction to deep learning. In the first lecture, we will introduce the concept of perceptron, which is the basis of deep learning and neural networks, through a linear regression task. Hope you enjoy it.

This lecture is the beginning of a series of lectures about Deep Learning. I do not plan to mention traditional machine learning, and if you are interested in it, please refer to the Machine Learning lectures by Prof. Andrew Ng.

I will introduce some experiments mainly based on PyTorch with this series. But there is almost nothing to experiment in the beginning, so this time there is only one Lab on Keras as an example, while Keras is more suitable for beginners BUT NOT FOR RESEARCHERS.

Linear Model

Firstly, we will review the unary linear regression and introduce the concept of gradient descent.