Artificial Intelligence (AI) & Data Mining
- Reinforcement Learning
- Neural learning algorithms, Slow Feature Analysis (SFA)
- Evolutionary strategies, neuroevolution
- Feature selection & feature construction
- Forecasting & time series analysis
- Random Forest
- Support Vector Machine (SVM)
Image Analysis & SIgnal Processing
- Pattern recognition & face recognition,
- Computational geometry (3D modeling, ASM/AAM),
- Image mosaicing,
- 3D Natural User Interfaces (NUIs): Kinect, Wii
- Medical image analysis (3D navigation).
Optimization, Simulation, Games
- Game learning, General game playing
- Simulation,
- Game Physics,
- Optimization: multi-criteria, EGO, Kriging.
Mini Glossary
| ASM | Active Shape Models. LV: WPF MIAV (BA) und WPF MIAV (MA). external links: www.isbe.man.ac.uk/~bim, http://www2.imm.dtu.dk/~aam. |
| AAM | Active Appearance Models.
LV: WPF MIAV (BA) und WPF MIAV (MA). |
| Slow Feature Analysis (SFA) | SFA is an analysing tool from neuroinformatics LV: Case Studies SOMA, BA-/MA-Theses external link: www.scholarpedia.org/article/Slow_feature_analysis (Laurenz Wiskott, Ruhr-Uni Bochum) |
| Reinforcement Learning (RL) | RL is an approach from Artificial Intelligence to solve optimization problems or control problems. The goal is to find a policy that steers an agent optimally in a given environment. The agent often gets feedback only after a number of steps (e.g. reward or punishment at the end of an episode). A famous early and successful application of RL is TD-Gammon, an agent developed by G. Tesauro, that learned with the help of RL to play the game Backgammon on championship level. LV: BA-/MA-Theses |
| (… to be continued …) |
LV: Teaching course which deepens this topic
