Sign inSign up
Achieving All with No Parameters: Adaptive NormalHedge thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Achieving All with No Parameters: Adaptive NormalHedge

Published on Aug 20, 20153092 Views

Haipeng Luo

We study the classic online learning problem of predicting with expert advice, and propose a truly parameter-free and adaptive algorithm that achieves several objectives simultaneously without using a

COLT 2015 - Paris

Related categories

Active LearningComputational Learning TheoryOn-line LearningReinforcement LearningSemi-supervised Learning

Chapter list

Achieving All with No Parameters: AdaNormalHedge00:00
How to Use Expert Advice?00:08
Hedge/Expert Setting00:24
Beyond the Classic Regret01:10
AdaNormalHedge01:54
Regret Guarantee - 103:05
Regret Guarantee - 203:28
Extensions and Applications04:04