VYSTADIAL - Development of statistical methods for spoken dialogue systems
The goals of the project is to study and improve statistical methods for learning of statistical models used in complex dialogue systems. The project will focus on development of robust learning techniques which will not be sensitive to the data sparsity problem. The data sparsity problem is especially typical in languages with rich morphology and syntax such as Czech. In addition, the project will target development of the advanced learning methods for statistical dialogue systems, including off-policy reinforcement learning methods.
MOBAme - Course development: Modern Bayesian methods
The goal of this project is to introduce a new course aimed at modern Bayesian inference methods. The course will emphasize and discuss methods which have application in robotics, natural language processing, data mining, and web search. The course will be composed of a series of lectures presented by experts from the Machine Learning Group, Cambridge University, UK, led by Prof. Zoubin Ghahramani. The course will be presented in English and it will be based on the machine learning course taught by Carl Edward Rasmussen and Zoubin Ghahramani at Cambridge University Engineering Department.
Alex - dialogue systems framework