Sciweavers

Share
ICML
2009
IEEE
10 years 8 months ago
Good learners for evil teachers
We consider a supervised machine learning scenario where labels are provided by a heterogeneous set of teachers, some of which are mediocre, incompetent, or perhaps even malicious...
Ofer Dekel, Ohad Shamir
ICML
2009
IEEE
10 years 8 months ago
Learning when to stop thinking and do something!
An anytime algorithm is capable of returning a response to the given task at essentially any time; typically the quality of the response improves as the time increases. Here, we c...
Barnabás Póczos, Csaba Szepesv&aacut...
ICML
2009
IEEE
10 years 8 months ago
ABC-boost: adaptive base class boost for multi-class classification
We propose abc-boost (adaptive base class boost) for multi-class classification and present abc-mart, an implementation of abcboost, based on the multinomial logit model. The key ...
Ping Li
ICML
2009
IEEE
10 years 8 months ago
Incorporating domain knowledge into topic modeling via Dirichlet Forest priors
Users of topic modeling methods often have knowledge about the composition of words that should have high or low probability in various topics. We incorporate such domain knowledg...
David Andrzejewski, Xiaojin Zhu, Mark Craven
ICML
2009
IEEE
10 years 8 months ago
Uncertainty sampling and transductive experimental design for active dual supervision
Dual supervision refers to the general setting of learning from both labeled examples as well as labeled features. Labeled features are naturally available in tasks such as text c...
Vikas Sindhwani, Prem Melville, Richard D. Lawrenc...
ICML
2009
IEEE
10 years 8 months ago
Compositional noisy-logical learning
We describe a new method for learning the conditional probability distribution of a binary-valued variable from labelled training examples. Our proposed Compositional Noisy-Logica...
Alan L. Yuille, Songfeng Zheng
ICML
2009
IEEE
10 years 8 months ago
Importance weighted active learning
We propose an importance weighting framework for actively labeling samples. This technique yields practical yet sound active learning algorithms for general loss functions. Experi...
Alina Beygelzimer, Sanjoy Dasgupta, John Langford
ICML
2009
IEEE
10 years 8 months ago
Route kernels for trees
Almost all tree kernels proposed in the literature match substructures without taking into account their relative positioning with respect to one another. In this paper, we propos...
Fabio Aiolli, Giovanni Da San Martino, Alessandro ...
ICML
2009
IEEE
10 years 8 months ago
Regularization and feature selection in least-squares temporal difference learning
We consider the task of reinforcement learning with linear value function approximation. Temporal difference algorithms, and in particular the Least-Squares Temporal Difference (L...
J. Zico Kolter, Andrew Y. Ng
books