Active learning for logistic regression: an evaluation

By Andrew I. Schein. 2007

356 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

Citations in the corpus

Active Bias: Training More Accurate Neural Networks by Emphasizing High Variance Samples (2017, methodology): “In the active learning setting, the prediction variance can be used to measure the uncertainty of each sample for either a regression or classification problem [41].”
Guest Editorial Active Learning and Intrinsically Motivated Exploration in Robots: Advances and Challenges (2010, background): “There have been active-extensions to most of the existing learning methods, e.g., logistic regression [12], support vector machines [13], and gaussian processes [14]–[16].”
Active Learning for Coreference Resolution (2012, background): “One common approach to assessing the informativeness is uncertainty sampling (Lewis and Gale, 1994; Schein and Ungar, 2007), in which the learner requests a label for the instance it is most uncertain how to label.”
Are All Training Examples Created Equal? An Empirical Study (2018, background): “However, there is evidence that it may not always work, and there are situations where random sampling performs better than some active learning algorithms [17, 5].”
Active learning of visual descriptors for grasping using non-parametric smoothed beta distributions (2012, background): “Active strategies for this type of techniques have been studied in [31] and pointed out a trade off between computational cost for experimental design based techniques and robustness issues that appear in more heuristic criteria.”
On The Power of Curriculum Learning in Training Deep Networks (2019, background): “…Learning (SPL- Kumar et al., 2010), boosting (Freund et al., 1996), hard example mining (Shrivastava et al., 2016) and even active learning (Schein & Ungar, 2007), mini-batches are sampled at each time point based on the ranking of the training examples by their difficulty with respect to…”
Generative Adversarial Active Learning (2017, unknown): “We use linear SVM as our classi-ﬁer although other classiﬁers can be used in active learning as well(Tong and Koller, 2002; Schein and Ungar, 2007; Settles, 2010).”
Active Learning Literature Survey (2009, background): “Schein and Ungar (2007) also show that active learning can sometimes require more labeled instances than “passive” supervised learning for the same model class, in their case logistic regression.”
Active Learning with Sampling by Uncertainty and Density for Word Sense Disambiguation and Text Classification (2008, methodology): “The Comp2 data set consists of comp.graphics and comp.windows.x categories from NewsGroups, which has been previously used in active learning for TC (Roy and McCallum, 2001; Schein and Ungar, 2007).”
Active Learning and Intrinsically Motivated Exploration in Robots : Advances and Challenges (2010, background): “There have been active-extensions to most of the existing learning methods, e.g., logistic regression [12], support vector machines [13], and gaussian processes [14]–[16].”
A Bayesian Approach for Policy Learning from Trajectory Preference Queries (2012, background): “This is not the first time active selection procedures have shown performance similar to passive methods [16].”