17 search results - page 4 / 4 » colt 1999 |
163
click to vote
COLT
1999
Springer
15 years 10 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...
145
click to vote
COLT
1999
Springer
15 years 10 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...