From the course: Deep Learning: Getting Started
Join today to access over 25,900 courses taught by industry experts.
Neural network architectures
From the course: Deep Learning: Getting Started
Neural network architectures
A neural network needs a blueprint before it can learn from data. This blueprint defines the overall design of the model. The model is built using layers and nodes. The architecture decides how these components are organized and connected. Once data enters the network, it moves through the design step by step. Each layer transforms the data before passing it forward. Not every problem benefits from the same design. A model built for images may look very different from a model built for language. CNNs, RNNs, LSTMs, and Transformers are common architecture types. Each one became popular because it solved a specific class of problems well. A simple neural network is a good starting point for basic predictions. But it may not capture complex patterns in specialized data. Different kinds of data carry information in different ways. For example, the position of pixels matters in images, while word order matters in language. Because of this, the model design must match the data. The wrong…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.
