SFB 876/C4 - Regression Approaches for Large-Scale High-Dimensional Data
- Prof. Dr. Katja Ickstadt
- Prof. Dr. Christian Sohler
Research supported by Deutsche Forschungsgemeinschaft, grant NA.
The scalability of modern regression approaches is often stretched to its limits by a large number of observations and/or variables. This aggravates their use in embedded systems. The goal of this project is therefore the development of highly efficient regression methods. We pursue the development of algorithms to reduce the number of observations using, e.g., random linear projections and sampling (streaming algorithms), as well as the development of methods to reduce the dimensionality of the underlying, possibly Bayesian, model classes imposing structural constraints, e.g., monotonicity.
Letzte Änderung am 20.11.2010 von A. Munteanu