Part 1: General introduction to AI & ML
- What are AI & ML and what can they do?
- What are the applications of AI & ML?
- What are the 3 main types of AI & ML learning and how to design a learning system?
- What are the important points regarding Training vs Test Distribution?
- What are the different function representations and search/optimization algorithms?
- What are the different metrics used to control the quality of the predictions?
Part 2: Data Wrangling
- Why is data pre-processing necessary in AI/ML?
- What data pre-processing steps need to be taken before building a model? – Data cleaning, data integration and transformation, data reduction, discretization and concept hierarchy generation.
Part 3: Performance Measures
- Weighted (Cost-Sensitive) Accuracy
- Precision/Recall (F-Measure, Breaking Even Point)
- Receiver Operating Characteristic (ROC) Curve (ROC Area)
Part 4: Neural Networks
- What is the inspiration to build neural networks?
- What are the different functions of an artificial neuron?
- What are the different types, topologies and properties of neural networks?
- What is the typical ML algorithm?
- What are the Encoder/Decoder, as well as Transformer?
- What are the applications of neural networks?
Part 5: Case Study
discussing how AI & ML can be used in Protein Engineering on the example of Complementary Factor I (CFI) protein
Check also our ML & AI Case Studies and other Resources: