Website introduction
“Dive into Deep Learning (D2L)” is an open source deep learning textbook written by Amazon scientist Li Mu and others. It is currently the most systematic tutorial suitable for introductory deep learning and has been adopted as a teaching material by 70+ universities around the world.
Deep learning is the core technology behind AI products such as ChatGPT and Midjourney. Understanding these technical principles will not only help you better use AI tools, but also allow you to participate in the development and innovation of AI applications.
Core learning content
- Deep learning basics：Linear regression, Softmax, multi-layer perceptron
- Convolutional Neural Network (CNN):Image recognition and computer vision
- Recurrent Neural Network (RNN): Natural Language Processing and Sequence Modeling
- Attention mechanism and Transformer: Understand the core architecture of GPT and BERT
- Generative Adversarial Network (GAN): Image generation principle
- Optimization algorithm: Gradient descent, Adam and other training techniques
Why recommended
Each chapter has executable PyTorch/TensorFlow code, and the effect is better when paired with Li Mu’s B-site video explanation. This is the best systematic learning path to understand how AI works, and it is also a key step to advance from an AI user to an AI developer.
Suitable for whom
- Developers and students who want to systematically learn the principles of deep learning
- AI practitioners who want to understand the principles of large model technologies such as ChatGPT
- Transformers who have a basic knowledge of Python and want to enter the AI industry