Sciweavers

13 search results - page 1 / 3
» Active Learning from Multiple Noisy Labelers with Varied Cos...
Sort
View
ICDM
2010
IEEE
228views Data Mining» more  ICDM 2010»
13 years 28 days ago
Active Learning from Multiple Noisy Labelers with Varied Costs
In active learning, where a learning algorithm has to purchase the labels of its training examples, it is often assumed that there is only one labeler available to label examples, ...
Yaling Zheng, Stephen D. Scott, Kun Deng
IJCV
2011
264views more  IJCV 2011»
12 years 10 months ago
Cost-Sensitive Active Visual Category Learning
Abstract We present an active learning framework that predicts the tradeoff between the effort and information gain associated with a candidate image annotation, thereby ranking un...
Sudheendra Vijayanarasimhan, Kristen Grauman
CVPR
2009
IEEE
14 years 10 months ago
What's It Going to Cost You?: Predicting Effort vs. Informativeness for Multi-Label Image Annotations
Active learning strategies can be useful when manual labeling effort is scarce, as they select the most informative examples to be annotated first. However, for visual category ...
Sudheendra Vijayanarasimhan (University of Texas a...
SIGIR
2011
ACM
12 years 6 months ago
Learning to rank from a noisy crowd
We study how to best use crowdsourced relevance judgments learning to rank [1, 7]. We integrate two lines of prior work: unreliable crowd-based binary annotation for binary classi...
Abhimanu Kumar, Matthew Lease
ICANN
2011
Springer
12 years 7 months ago
Learning from Multiple Annotators with Gaussian Processes
Abstract. In many supervised learning tasks it can be costly or infeasible to obtain objective, reliable labels. We may, however, be able to obtain a large number of subjective, po...
Perry Groot, Adriana Birlutiu, Tom Heskes