COLT
2010
Springer
13 years 2 months ago
2010
Springer
In standard online learning, the goal of the learner is to maintain an average loss that is "not too big" compared to the loss of the best-performing function in a fixed...
ICMLA
2004
13 years 6 months ago
2004
Landmarking is a recent and promising metalearning strategy, which defines meta-features that are themselves efficient learning algorithms. However, the choice of landmarkers is m...
COLT
2006
Springer
13 years 8 months ago
2006
Springer
We describe a new approach for understanding the difficulty of designing efficient learning algorithms. We prove that the existence of an efficient learning algorithm for a circui...
ICML
2000
IEEE
14 years 5 months ago
2000
IEEE
Landmarking is a novel approach to describing tasks in meta-learning. Previous approaches to meta-learning mostly considered only statistics-inspired measures of the data as a sou...
CVPR
2007
IEEE
14 years 6 months ago
2007
IEEE
Many low-level vision algorithms assume a prior probability over images, and there has been great interest in trying to learn this prior from examples. Since images are very non G...