17 search results - page 4 / 4 » colt 1999 |
COLT
1999
Springer
13 years 9 months ago
1999
Springer
We present a unified framework for reasoning about worst-case regret bounds for learning algorithms. This framework is based on the theory of duality of convex functions. It brin...
COLT
1999
Springer
13 years 9 months ago
1999
Springer
The empirical error on a test set, the hold-out estimate, often is a more reliable estimate of generalization error than the observed error on the training set, the training estim...