Overview
Semester | Summer 2023 |
ECTS | 4 |
Level | Master |
Description
The course will be held in English.
Reproducing (scientific) results is more important than ever before because of the impact of AI systems these days. This applies to researchers to validate existing insights, but also to developers using approaches from papers in their products. The goal of this lab is to read state-of-the-art papers in the field of ML and reproduce their results. This can include running existing software packages, comparing results of different implementations or even reimplementing ideas from scratch. So, the course will enable you to read, understand and implement research papers, i.e. a skill set required in most AI-driven companies these days.
Recommended pre-requisites
- Machine Learning
- Deep Learning
Lecturer


30167 Hannover


Topics
- Hyperparameter Optimization
- Neural Architecture Search
- Meta-Learning
- Dynamic Configuration
- Algorithm Selection
Literature
There is no recommended literature for this course.