Study Guide

Objective

You understand how neural networks work and know how to use them in practical machine learning work
You can apply different types of neural networks for deep learning applications and different use cases
You can create suitable error metrics for your deep learning model
You can pre-process data in a suitable way for your deep learning models
You can share your results and exercises via a version control system.

Content

Basics of deep learning and how to design and optimize neural networks
Deep learning vs. classical machine learning
Basics of neural networks: weights, biases, hidden layers, neurons/nodes, activation function, loss function, optimizers, training and inference
ANN/MLP, CNN, RNN and LSTM networks
Basics of transformer networks

Qualifications

Basics of programming, Basics of conventional machine learning methods, Basics of Python data analytics modules/libraries

Assessment criteria, satisfactory (1)

You can create deep learning applications with limited features
You are aware of the concepts of how neural networks work in practice
You can do minimal pre-processing for data regarding deep learning applications
You can create basic error metrics for your deep learning models
You can share your results and exercises via a version control system.

Assessment criteria, good (3)

You can create various deep learning applications
You can experiment with the concepts of how neural networks work in practice by using code
You can do suitable pre-processing for data regarding deep learning applications
You can create suitable error metrics for your deep learning models
You can share your results and exercises via a version control system.

Assessment criteria, excellent (5)

You can create various deep learning applications
You can experiment with the concepts of how neural networks work in practice by using code
You can do suitable pre-processing for data regarding deep learning applications
You can create suitable error metrics for your deep learning models
You can optimize your deep learning models and design suitable neural network structures for your data
You can share your results and exercises via a version control system.