Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

8

TIT
2008

favoriteEmaildiscussreport

76views more TIT 2008»

Improved Risk Tail Bounds for On-Line Algorithms

13 years 4 months ago

Improved Risk Tail Bounds for On-Line Algorithms

Download books.nips.cc

We prove the strongest known bound for the risk of hypotheses selected from the ensemble generated by running a learning algorithm incrementally on the training data. Our result is based on proof techniques that are remarkably different from the standard risk analysis based on uniform convergence arguments.

Nicolò Cesa-Bianchi, Claudio Gentile

Real-time Traffic

Proof Techniques | Standard Risk Analysis | TIT 2008 | Uniform Convergence Arguments |

claim paper

Related Content

» On the Generalization Ability of OnLine Learning Algorithms

» Improved OnLine Broadcast Scheduling with Deadlines

» An OnLine Algorithm for Improving Performance in Navigation

» Efficient OnLine Frequency Allocation and Call Control in Cellular Networks

» Spaceoptimal heavy hitters with strong error bounds

» An Efficient Motion Planning Algorithm for Stochastic Dynamic Systems with Constraints on ...

» Online Least Squares Estimation with SelfNormalized Processes An Application to Bandit Pro...

» DRRSFF A Practical Scheduling Algorithm to Improve the Performance of Short Flows

» A Randomized Online Learning Algorithm for Better Variance Control

Post Info
More Details (n/a)

Added	15 Dec 2010
Updated	15 Dec 2010
Type	Journal
Year	2008
Where	TIT
Authors	Nicolò Cesa-Bianchi, Claudio Gentile

Comments (0)