Data-centric Artificial Intelligence

Aim

After completing the course, the student should be able to:

• explain and apply basic concepts in data-centric artificial intelligence
• explain and systematically reflect on the applicability of different methods in data-centric artificial intelligence for different types of learning problems
• use existing tools to construct, train, and analyze deep neural network models
• develop and adapt code to preprocess, analyze and improve the quality of data sets

Content

The course provides basic methodological knowledge in data-centric artificial intelligence through the following content:

• basic concepts in artificial neural networks for deep learning
• methods and tools for construction and adaptation, training, and evaluation of models (like Python, Keras, Tensorflow, Colab)
• data-centric methods for detailed error analysis of models and underlying data
• structured approach to managing and correcting (curating) data