Combining model-based and instance-based learning for first order regression

T ORDER REGRESSION (EXTENDED ABSTRACT) Kurt Driessensa Saso Dzeroskib a Department of Computer Science, University of Waikato, Hamilton, New Zealand ([e-mail]) b Department of Knowledge Technologies, Jozef Stefan Institute, Ljubljana, Slovenia ([e-mail]) The full paper on this topic appears in the Proceedings of the 22nd International Conference on Machine Learning, Bonn, Germany, 2005. [1] With the development of relational reinforcement learning [5, 4] came the need for incremental relational regression algorithms. A relational regression algorithm generalizes over learning examples with a continuous target value and makes predictions about the value of unseen examples, using a relational representation for both the learning examples and the resulting function. A number of these algorithms have already been developed. The tg algorithm builds regression trees [3], rib uses instance based regression with first order distances [2] and kbr relies on Gaussian proces...

• Type: Conference
• Year: 2005
• Where: ICML
• Authors: Kurt Driessens, Saso Dzeroski
• Keywords: Algorithm Builds Regression, ICML 2005, Learning Examples, Machine Learning, Relational Regression Algorithms