Scope

This workshop seeks to bring together the participants’ expertise in machine learning to achieve a better understanding of the problem and devise novel approaches to transfer learning. In addition to new approaches, applications and evaluation metrics, there is also a need to develop consensus on standardized domains and testbeds and identify new subproblems and representational choices.  We  welcome submissions from experimentalists, theorists, and psychological modelers with contributions that deal with transfer in classification, inference, reactive control, and planning tasks. We also welcome work cast in logical, probabilistic, and other theoretical frameworks. Initial attempts at transfer learning have suggested that knowledge representation plays a crucial role in any form of transfer. We particularly seek contributions from logic and KR communities to better understand the conditions under which successful and useful transfer can be expected. We also seek to learn from the experiences of cognitive and computational psychologists.

Attendance policy