- Brain-Computer interface
- Motion classification and anticipation from EEG and EMG signals for virtual reality applications and for technological aids for the neuro-motor impaired
- Function estimation by Support Vector Machines
- Random embedding and boosting machines for pattern recognition
- Clustering and unsupervised learning by Bayesian methods.
- Support vector machines for clustering and pattern recognition.
- Augmented (real+virtual) reality for aiding people with movement disorders. (there is a special fellowship for this).
- Information geometric methods for sampling, learning and optimization.
- Collision-free traffic control by neural networks.
- Process prediction (e.g., financial forcasting) by nonlinear methods and neural networks.
Please contact Prof. Baram for details.