Active Learning Query Strategies Research Articles

Retracted: Active Learning Query Strategies for Linear Regression Based on Efficient Global Optimization

Retracted: Active Learning Query Strategies for Linear Regression Based on Efficient Global Optimization

Onception: Active Learning with Expert Advice for Real World Machine Translation

Abstract Active learning can play an important role in low-resource settings (i.e., where annotated data is scarce), by selecting which instances may be more worthy to annotate. Most active learning approaches for Machine Translation assume the existence of a pool of sentences in a source language, and rely on human annotators to provide translations or post-edits, which can still be costly. In this article, we apply active learning to a real-world human-in-the-loop scenario in which we assume that: (1) the source sentences may not be readily available, but instead arrive in a stream; (2) the automatic translations receive feedback in the form of a rating, instead of a correct/edited translation, since the human-in-the-loop might be a user looking for a translation, but not be able to provide one. To tackle the challenge of deciding whether each incoming pair source–translations is worthy to query for human feedback, we resort to a number of stream-based active learning query strategies. Moreover, because we do not know in advance which query strategy will be the most adequate for a certain language pair and set of Machine Translation models, we propose to dynamically combine multiple strategies using prediction with expert advice. Our experiments on different language pairs and feedback settings show that using active learning allows us to converge on the best Machine Translation systems with fewer human interactions. Furthermore, combining multiple strategies using prediction with expert advice outperforms several individual active learning strategies with even fewer interactions, particularly in partial feedback settings.

Toward Label-Efficient Neural Network Training: Diversity-Based Sampling in Semi-Supervised Active Learning

Collecting large-labeled data is an expensive and challenging issue for training deep neural networks. To address this issue, active learning is recently studied where the active learner provides informative samples for labeling. Diversity-based sampling algorithms are commonly used for representation-based active learning. In this paper, a new diversity-based sampling is introduced for semi-supervised active learning. To select more informative data at the initial stage, we devise a diversity-based initial dataset selection method by using self-supervised representation. We further propose a new active learning query strategy, which exploits both consistency and diversity. Comparative experiments show that the proposed method can outperform other active learning approaches on two public datasets.

A framework to build accurate Convolutional Neural Network models for melanoma diagnosis

In the past few years, Convolutional Neural Networks have achieved performance levels similar to those achieved by dermatologists. However, the diagnosis of melanoma remains a challenging task, mainly due to the high levels of inter and intra-class variability present in images of moles. With the aim of new methods for an effective melanoma diagnosis, a new framework is proposed. The training process is guided by an expert within an active learning approach where the architectures implicitly learn about the complexity of individual images through query strategies, which allows us to adjust the training process and achieve better performance. In addition, we propose a batch-based query strategy that enables a more stable and faster training process. Besides, the framework leverages segmentation, data augmentation and transfer learning to enhance melanoma diagnosis. The framework is composed by several specialized blocks, which allow us to measure how the diagnosis is improved after each step. In this sense, blocks could be customized and do not depend on specific models. An extensive experimental study was conducted on 16 skin image datasets, where five state-of-the-art models were significantly outperformed. This study corroborated that new active learning query strategies can be employed to effectively train neural networks architectures for the diagnosis of melanoma, achieving 182% better predictive performance in Xception, and an overall 11% and 20% better predictive performance in dermoscopic and non-dermoscopic images, respectively. It is worth mentioning that the informativeness value of each image is shown, which leads to identify the hardest images for the predictive models. Finally, the proposal required 2% of the total training time, and needed 61% less training epochs.

On the application of active learning for efficient and effective IoT botnet detection

The active learning approach for machine learning can greatly benefit those environments where a wealth of unlabeled data is available, and the labeling cost of the data can be restrictive. In this regard, security operations centers (SOCs) can take advantage of the human expertise available to improve machine learning-based detection models using the active learning approach. In the context of SOC operations and IoT botnet detection, our study provides a thorough benchmarking of the application of different active learning approaches within the framework of pool-based sampling. The selection of the optimal query instance for learning is evaluated using uncertainty sampling, ranked batch-mode sampling, and query by committee strategies. Our results show that the active learning approach can help to generate better detection models using all the active learning query strategies tested in our benchmarking setup. Leveraging the human–machine interaction can produce high-performance models in the context of IoT botnet detection using significantly less data than the passive approaches traditionally used for the generation of machine learning-based detection systems. Additionally, the impact of wrong-labeled data in the active learning implementation is explored.

Active Learning Query Strategies for Linear Regression Based on Efficient Global Optimization

Active learning, a subfield of machine learning, can train a good model by selecting a minimum number of labeled samples. In many machine learning scenarios, needed information (such as the best value in unlabeled datasets) is acquired by prediction. When there is too little data in the training model, the prediction accuracy would obviously affect the accuracy of the results. To establish a high-performance regression model for a small dataset while accelerating the search for the best sample, a new active learning query strategy, EGO-ALR, that combines efficient global optimization (EGO) and active learning for regression (ALR) was proposed. It was found that the performance of EGO-ALR was significantly better than the original ALR query strategies in terms of the root mean square error (RMSE), correlation coefficient (CC), and opportunity cost (Oppo Cost). Specifically, EGO-ALR increased the Oppo Cost by an average of 25.27% when the RMSE and CC values were not more than 1.07% different from the original ALR. This study validated the efficiency and robustness of EGO-ALR approaches using 19 datasets from various domains and three distinct linear regression models (Ridge regression, Lasso, and Elastic network).

Zero Initialized Active Learning with Spectral Clustering using Hungarian Method

Supervised machine learning tasks often require a large number of labeled training data to set up a model, and then prediction - for example the classification - is carried out based on this model. Nowadays tremendous amount of data is available on the web or in data warehouses, although only a portion of those data is annotated and the labeling process can be tedious, expensive and time consuming. Active learning tries to overcome this problem by reducing the labeling cost through allowing the learning system to iteratively select the data from which it learns. In special case of active learning, the process starts from zero initialized scenario, where the labeled training dataset is empty, and therefore only unsupervised methods can be performed. In this paper a novel query strategy framework is presented for this problem, called Clustering Based Balanced Sampling Framework (CBBSF), which is not only select the initial labeled training dataset, but uniformly selects the items among the categories to get a balanced labeled training dataset. The framework includes an assignment technique to implicitly determine the class membership probabilities. Assignment solution is updated during CBBSF iterations, hence it simulates supervised machine learning more accurately as the process progresses. The proposed Spectral Clustering Based Sampling (SCBS) query startegy realizes the CBBSF framework, and therefore it is applicable in the special zero initialized situation. This selection approach uses ClusterGAN (Clustering using Generative Adversarial Networks) integrated in the spectral clustering algorithm and then it selects an unlabeled instance depending on the class membership probabilities. Global and local versions of SCBS were developed, furthermore, most confident and minimal entropy measures were calculated, thus four different SCBS variants were examined in total. Experimental evaluation was conducted on the MNIST dataset, and the results showed that SCBS outperforms the state-of-the-art zero initialized active learning query strategies.

A Comparative Analysis of Active Learning for Biomedical Text Mining

An enormous amount of clinical free-text information, such as pathology reports, progress reports, clinical notes and discharge summaries have been collected at hospitals and medical care clinics. These data provide an opportunity of developing many useful machine learning applications if the data could be transferred into a learn-able structure with appropriate labels for supervised learning. The annotation of this data has to be performed by qualified clinical experts, hence, limiting the use of this data due to the high cost of annotation. An underutilised technique of machine learning that can label new data called active learning (AL) is a promising candidate to address the high cost of the label the data. AL has been successfully applied to labelling speech recognition and text classification, however, there is a lack of literature investigating its use for clinical purposes. We performed a comparative investigation of various AL techniques using ML and deep learning (DL)-based strategies on three unique biomedical datasets. We investigated random sampling (RS), least confidence (LC), informative diversity and density (IDD), margin and maximum representativeness-diversity (MRD) AL query strategies. Our experiments show that AL has the potential to significantly reducing the cost of manual labelling. Furthermore, pre-labelling performed using AL expediates the labelling process by reducing the time required for labelling.

Investigation of Combining Logitboost(M5P) under Active Learning Classification Tasks

Active learning is the category of partially supervised algorithms that is differentiated by its strategy to combine both the predictive ability of a base learner and the human knowledge so as to exploit adequately the existence of unlabeled data. Its ambition is to compose powerful learning algorithms which otherwise would be based only on insufficient labelled samples. Since the latter kind of information could raise important monetization costs and time obstacles, the human contribution should be seriously restricted compared with the former. For this reason, we investigate the use of the Logitboost wrapper classifier, a popular variant of ensemble algorithms which adopts the technique of boosting along with a regression base learner based on Model trees into 3 different active learning query strategies. We study its efficiency against 10 separate learners under a well-described active learning framework over 91 datasets which have been split to binary and multi-class problems. We also included one typical Logitboost variant with a separate internal regressor for discriminating the benefits of adopting a more accurate regression tree than one-node trees, while we examined the efficacy of one hyperparameter of the proposed algorithm. Since the application of the boosting technique may provide overall less biased predictions, we assume that the proposed algorithm, named as Logitboost(M5P), could provide both accurate and robust decisions under active learning scenarios that would be beneficial on real-life weakly supervised classification tasks. Its smoother weighting stage over the misclassified cases during training as well as the accurate behavior of M5P are the main factors that lead towards this performance. Proper statistical comparisons over the metric of classification accuracy verify our assumptions, while adoption of M5P instead of weak decision trees was proven to be more competitive for the majority of the examined problems. We present our results through appropriate summarization approaches and explanatory visualizations, commenting our results per case.

Multi-category Classification Problem Oriented Subsampling-Based Active Learning Method

Traditional active learning methods have achieved gratifying results in the classification tasks of less categories such as binary classification, the application research of active learning in the field of big data problems still faces enormous challenges. Since many active learning query strategies need to perform matrix inversion, the amount of calculation increases exponentially with the increase of the scale of the problem, it is difficult to apply these active learning methods in large scale multi-category data classification task. In order to solve this problem, this paper designed a subsampling-based active learning model, and integrate unsupervised clustering algorithm with traditional active learning method, then conducted experiments on Binary Alphadigits and OMNIGLOT data sets. This paper compares the performance of five traditional active learning algorithms using this subsampling method, namely random sampling, uncertainty sampling, query-by-committee, density weighting and learning-based active learning. Through comparative experiments, the feasibility of active learning based on subsampling for solving the multi-category classification problem is verified, and it is found that the subsampling-based method can break the limitations of traditional active learning methods that cannot deal with large-scale data classification.

SEAL: Semisupervised Adversarial Active Learning on Attributed Graphs.

Active learning (AL) on attributed graphs has received increasing attention with the prevalence of graph-structured data. Although AL has been widely studied for alleviating label sparsity issues with the conventional nonrelational data, how to make it effective over attributed graphs remains an open research question. Existing AL algorithms on node classification attempt to reuse the classic AL query strategies designed for nonrelational data. However, they suffer from two major limitations. First, different AL query strategies calculated in distinct scoring spaces are often naively combined to determine which nodes to be labeled. Second, the AL query engine and the learning of the classifier are treated as two separating processes, resulting in unsatisfactory performance. In this article, we propose a SEmisupervised Adversarial active Learning (SEAL) framework on attributed graphs, which fully leverages the representation power of deep neural networks and devises a novel AL query strategy for node classification in an adversarial way. Our framework learns two adversarial components; a graph embedding network that encodes both the unlabeled and labeled nodes into a common latent space, expecting to trick the discriminator to regard all nodes as already labeled, and a semisupervised discriminator network that distinguishes the unlabeled from the existing labeled nodes. The divergence score, generated by the discriminator in a unified latent space, serves as the informativeness measure to actively select the most informative node to be labeled by an oracle. The two adversarial components form a closed loop to mutually and simultaneously reinforce each other toward enhancing the AL performance. Extensive experiments on real-world networks validate the effectiveness of the SEAL framework with superior performance improvements to state-of-the-art baselines on node classification tasks.

Active Learning Query Strategies for Classification, Regression, and Clustering: A Survey

Generally, data is available abundantly in unlabeled form, and its annotation requires some cost. The labeling, as well as learning cost, can be minimized by learning with the minimum labeled data instances. Active learning (AL), learns from a few labeled data instances with the additional facility of querying the labels of instances from an expert annotator or oracle. The active learner uses an instance selection strategy for selecting those critical query instances, which reduce the generalization error as fast as possible. This process results in a refined training dataset, which helps in minimizing the overall cost. The key to the success of AL is query strategies that select the candidate query instances and help the learner in learning a valid hypothesis. This survey reviews AL query strategies for classification, regression, and clustering under the pool-based AL scenario. The query strategies under classification are further divided into: informative-based, representative-based, informative- and representative-based, and others. Also, more advanced query strategies based on reinforcement learning and deep learning, along with query strategies under the realistic environment setting, are presented. After a rigorous mathematical analysis of AL strategies, this work presents a comparative analysis of these strategies. Finally, implementation guide, applications, and challenges of AL are discussed.

Overlap Aware Active Learning Query Strategies for Pool Based Scenario

Active learning (AL) is an emerging machine learning technique, which works with fewer labeled instances, thereby reducing labeling as well as learning cost. Various AL query strategies are available to identify the candidate query instances for achieving a generalized hypothesis. However, the AL query strategies sometimes produce redundant or overlapped instances as candidate instances, due to the presence of instance and/or class overlapping in the dataset. This overlapping results in a complex decision boundary. We addressed this problem and proposed a solution as an augmented step for AL query strategies, which improves the generalization performance of the AL task. The solution is based on the nearest neighbor assumption and fixed vicinity based approach. This augmentation tries to avoid the selection of overlap instances and provides a smoother decision boundary. We have validated the performance of the proposed method on the available benchmark datasets.

On active learning methods for manifold data

Active learning is a major area of interest within the field of machine learning, especially when the labeled instances are very difficult, time-consuming or expensive to obtain. In this paper, we review various active learning methods for manifold data, where the intrinsic manifold structure of data is also incorporated into the active learning query strategies. In addition, we present a new manifold-based active learning algorithm for Gaussian process classification. This new method uses a data-dependent kernel derived from a semi-supervised model that considers both labeled and unlabeled data. The method performs a regularization on the smoothness of the fitted function with respect to both the ambient space and the manifold where the data lie. The regularization parameter is treated as an additional kernel (covariance) parameter and estimated from the data, permitting adaptation of the kernel to the given dataset manifold geometry. Comparisons with other AL methods for manifold data show faster learning performance in our empirical experiments. MATLAB code that reproduces all examples is provided as supplementary materials.

Active learning reduces annotation time for clinical concept extraction

ObjectiveTo investigate: (1) the annotation time savings by various active learning query strategies compared to supervised learning and a random sampling baseline, and (2) the benefits of active learning-assisted pre-annotations in accelerating the manual annotation process compared to de novo annotation. Materials and methodsThere are 73 and 120 discharge summary reports provided by Beth Israel institute in the train and test sets of the concept extraction task in the i2b2/VA 2010 challenge, respectively. The 73 reports were used in user study experiments for manual annotation. First, all sequences within the 73 reports were manually annotated from scratch. Next, active learning models were built to generate pre-annotations for the sequences selected by a query strategy. The annotation/reviewing time per sequence was recorded. The 120 test reports were used to measure the effectiveness of the active learning models. ResultsWhen annotating from scratch, active learning reduced the annotation time up to 35% and 28% compared to a fully supervised approach and a random sampling baseline, respectively. Reviewing active learning-assisted pre-annotations resulted in 20% further reduction of the annotation time when compared to de novo annotation. DiscussionThe number of concepts that require manual annotation is a good indicator of the annotation time for various active learning approaches as demonstrated by high correlation between time rate and concept annotation rate. ConclusionActive learning has a key role in reducing the time required to manually annotate domain concepts from clinical free text, either when annotating from scratch or reviewing active learning-assisted pre-annotations.

Low-Resource Active Learning of Morphological Segmentation

Many Uralic languages have a rich morphological structure, but lack morphological analysis tools needed for efficient language processing. While creating a high-quality morphological analyzer requires a significant amount of expert labor, data-driven approaches may provide sufficient quality for many applications. We study how to create a statistical model for morphological segmentation with a large unannotated corpus and a small amount of annotated word forms selected using an active learning approach. We apply the procedure to two Finno-Ugric languages: Finnish and North Sámi. The semi-supervised Morfessor FlatCat method is used for statistical learning. For Finnish, we set up a simulated scenario to test various active learning query strategies. The best performance is provided by a coverage-based strategy on word initial and final substrings. For North Sámi we collect a set of humanannotated data. With 300 words annotated with our active learning setup, we see a relative improvement in morph boundary F1-score of 19% compared to unsupervised learning and 7.8% compared to random selection.