Maximizing Expected Model Change for Active Learning in Regression

Cai, Wenli; Zhang, Ya; Zhou, Jun

doi:10.1109/icdm.2013.104

Cited by 139 publications

(115 citation statements)

References 11 publications

Supporting

Mentioning

109

Contrasting

Order By: Relevance

“…In this paper, we extend our previous work [18] to BMAL by simulating the sequential mode AL behavior to simultaneously choose a set of examples without retraining, which is the new contributions of this paper. Now, we present the formulations of our AL methods targeting on regression, and then extend them to BMAL algorithms.…”

Section: Batch Mode Active Learningmentioning

confidence: 93%

“…Yu and Kim [17] provided passive sampling heuristics based on the geometric characteristics of data. Cai et al [18] presented a novel data sampling solution in the context of regression, which queries the example leading to the largest model change.…”

Section: B Active Learning For Regressionmentioning

confidence: 99%

“…In this paper, we consider a hard AL scenario, where the size of initial training set is small: D(10%) + U(70%) + T (20%). For the features, we normalize the features as [18].…”

Section: A Data Sets and Experimental Setupsmentioning

confidence: 99%

See 2 more Smart Citations

Batch Mode Active Learning for Regression With Expected Model Change

Cai

Zhang

2017

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

While active learning (AL) has been widely studied for classification problems, limited efforts have been done on AL for regression. In this paper, we introduce a new AL framework for regression, expected model change maximization (EMCM), which aims at choosing the unlabeled data instances that result in the maximum change of the current model once labeled. The model change is quantified as the difference between the current model parameters and the updated parameters after the inclusion of the newly selected examples. In light of the stochastic gradient descent learning rule, we approximate the change as the gradient of the loss function with respect to each single candidate instance. Under the EMCM framework, we propose novel AL algorithms for the linear and nonlinear regression models. In addition, by simulating the behavior of the sequential AL policy when applied for k iterations, we further extend the algorithms to batch mode AL to simultaneously choose a set of k most informative instances at each query time. Extensive experimental results on both UCI and StatLib benchmark data sets have demonstrated that the proposed algorithms are highly effective and efficient.

show abstract

Section: Batch Mode Active Learningmentioning

confidence: 93%

Section: B Active Learning For Regressionmentioning

confidence: 99%

See 1 more Smart Citation

Batch Mode Active Learning for Regression With Expected Model Change

Cai

Zhang

2017

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…• Expected model change maximization [10]: This algorithm quantifies the change as the difference between the current model parameters and the new model parameters, and chooses an unlabeled instance which results in the greatest change.…”

Section: Methodsmentioning

confidence: 99%

A Graph-Based Approach for Active Learning in Regression

Zhang¹,

Ravi²,

Davidson³

2020

Proceedings of the 2020 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

Active learning aims to reduce labeling efforts by selectively asking humans to annotate the most important data points from an unlabeled pool and is an example of human-machine interaction. Though active learning has been extensively researched for classification and ranking problems, it is relatively understudied for regression problems. Most existing active learning for regression methods use the regression function learned at each active learning iteration to select the next informative point to query. This introduces several challenges such as handling noisy labels, parameter uncertainty and overcoming initially biased training data. Instead, we propose a feature-focused approach that formulates both sequential and batch-mode active regression as a novel bipartite graph optimization problem. We conduct experiments on both noise-free and noisy settings. Our experimental results on benchmark data sets demonstrate the effectiveness of our proposed approach.• We formulate a feature (not model) focused active learning algorithm as a bipartite graph opti-

show abstract

“…More generally, regarding the use of such hypothesis generation methods within a medical context, it is possible can regard the iterative process of experimental feedback and problem refinement as one of Active Learning [30]. Activelearning addresses the problem of how to choose the most informative examples to train a machine learner; it does so by selecting those examples that most effectively differentiate between hypotheses, thereby minimizing the number of experimental surveys that need to be conducted in order to form a complete model.…”

Section: Future Outlookmentioning

confidence: 99%

A Kernel-Based Framework for Medical Big-Data Analytics

Windridge

Bober

2014

Interactive Knowledge Discovery and Data Mining in Biomedical Informatics

View full text Add to dashboard Cite

Abstract. The recent trend towards standardization of Electronic Health Records (EHRs) represents a significant opportunity and challenge for medical big-data analytics. The challenge typically arises from the nature of the data which may be heterogeneous, sparse, very highdimensional, incomplete and inaccurate. Of these, standard pattern recognition methods can typically address issues of high-dimensionality, sparsity and inaccuracy. The remaining issues of incompleteness and heterogeneity however are problematic; data can be as diverse as handwritten notes, blood-pressure readings and MR scans, and typically very little of this data will be co-present for each patient at any given time interval.We therefore advocate a kernel-based framework as being most appropriate for handling these issues, using the neutral point substitution method to accommodate missing inter-modal data. For pre-processing of image-based MR data we advocate a Deep Learning solution for contextual areal segmentation, with edit-distance based kernel measurement then used to characterize relevant morphology.

show abstract

Maximizing Expected Model Change for Active Learning in Regression

Cited by 139 publications

References 11 publications

Batch Mode Active Learning for Regression With Expected Model Change

Batch Mode Active Learning for Regression With Expected Model Change

A Graph-Based Approach for Active Learning in Regression

A Kernel-Based Framework for Medical Big-Data Analytics

Contact Info

Product

Resources

About