COLT
2008
Springer
13 years 9 months ago
2008
Springer
We introduce an efficient algorithm for the problem of online linear optimization in the bandit setting which achieves the optimal O ( T) regret. The setting is a natural general...
COLT
2008
Springer
13 years 9 months ago
2008
Springer
COLT
2008
Springer
13 years 9 months ago
2008
Springer
COLT
2008
Springer
13 years 9 months ago
2008
Springer
COLT
2008
Springer
13 years 9 months ago
2008
Springer
Standard no-internal-regret (NIR) algorithms compute a fixed point of a matrix, and hence typically require O(n3 ) run time per round of learning, where n is the dimensionality of...
COLT
2008
Springer
13 years 9 months ago
2008
Springer
COLT
2008
Springer
13 years 9 months ago
2008
Springer
In the multi-view learning paradigm, the input variable is partitioned into two different views X1 and X2 and there is a target variable Y of interest. The underlying assumption i...
COLT
2008
Springer
13 years 9 months ago
2008
Springer
A kernel over the Boolean domain is said to be reflection-invariant, if its value does not change when we flip the same bit in both arguments. (Many popular kernels have this prop...
COLT
2008
Springer
13 years 9 months ago
2008
Springer
An algorithm is presented for online learning of rotations. The proposed algorithm involves matrix exponentiated gradient updates and is motivated by the von Neumann divergence. T...
COLT
2008
Springer
13 years 9 months ago
2008
Springer
We propose a model-based learning algorithm, the Adaptive Aggregation Algorithm (AAA), that aims to solve the online, continuous state space reinforcement learning problem in a de...