TY - JOUR
T1 - Learning with continuous experts using drifting games
AU - Mukherjee, Indraneel
AU - Schapire, Robert E.
N1 - Funding Information:
Thanks to Jake Abernethy and Yoav Freund for many helpful discussions. This research was supported by NSF grants IIS-0325500.
PY - 2008
Y1 - 2008
N2 - We consider the problem of learning to predict as well as the best in a group of experts making continuous predictions. We assume the learning algorithm has prior knowledge of the maximum number of mistakes of the best expert. We propose a new master strategy that achieves the best known performance for online learning with continuous experts in the mistake bounded model. Our ideas are based on drifting games, a generalization of boosting and online learning algorithms. We also prove new lower bounds based on the drifting games framework which, though not as tight as previous bounds, have simpler proofs and do not require an enormous number of experts.
AB - We consider the problem of learning to predict as well as the best in a group of experts making continuous predictions. We assume the learning algorithm has prior knowledge of the maximum number of mistakes of the best expert. We propose a new master strategy that achieves the best known performance for online learning with continuous experts in the mistake bounded model. Our ideas are based on drifting games, a generalization of boosting and online learning algorithms. We also prove new lower bounds based on the drifting games framework which, though not as tight as previous bounds, have simpler proofs and do not require an enormous number of experts.
UR - http://www.scopus.com/inward/record.url?scp=57149142047&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=57149142047&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-87987-9_22
DO - 10.1007/978-3-540-87987-9_22
M3 - Conference article
AN - SCOPUS:57149142047
SN - 0302-9743
VL - 5254 LNAI
SP - 240
EP - 255
JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
T2 - 19th International Conference on Algorithmic Learning Theory, ALT 2008
Y2 - 13 October 2008 through 16 October 2008
ER -