Advanced Deep Learning Techniques
In the previous chapter, we reviewed the concept of neural network modeling and deep learning while focusing on fully connected neural networks. In this chapter, we will discuss more advanced techniques that let you use deep learning models across different data types and structures, such as images, texts, and graphs. These techniques are behind the majority of advancements across industries through artificial intelligence, such as in chatbots, medical diagnosis, drug discovery, stock trading, and fraud detection. Although we will present some of the most famous deep learning models across different data types, this chapter aims to help you understand the concepts and practice with PyTorch, and not provide you with state-of-the-art models for each data type or subject domain.
In this chapter, we will cover the following topics:
- Types of neural networks
- Convolutional neural networks for image shape data
- Transformers for language modeling...
