Sciweavers

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

11

ICML
2007
IEEE

favoriteEmaildiscussreport

158views Machine Learning» more ICML 2007»

A bound on the label complexity of agnostic active learning

14 years 5 months ago

A bound on the label complexity of agnostic active learning

Download www.cs.cmu.edu

We study the label complexity of pool-based active learning in the agnostic PAC model. Specifically, we derive general bounds on the number of label requests made by the A2 algorithm proposed by Balcan, Beygelzimer & Langford (Balcan et al., 2006). This represents the first nontrivial general-purpose upper bound on label complexity in the agnostic PAC model.

Steve Hanneke

Real-time Traffic

Agnostic Pac Model | Balcan Et | ICML 2007 | Label Complexity | Machine Learning |

claim paper

Related Content

» A general agnostic active learning algorithm

» Complexity Bounds for Batch Active Learning in Classification

» Agnostic active learning

» Analysis of a greedy active learning strategy

» Lower Bounds for Agnostic Learning via Approximate Rank

» A Lower Bound for Agnostically Learning Disjunctions

» Active Learning in the Nonrealizable Case

» OptimallySmooth Adaptive Boosting and Application to Agnostic Learning

» Agnostic Learning versus Prior Knowledge in the Design of Kernel Machines

Post Info
More Details (n/a)

Added	17 Nov 2009
Updated	17 Nov 2009
Type	Conference
Year	2007
Where	ICML
Authors	Steve Hanneke

Comments (0)