Abstract Active Learning Research Articles

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.

Partitioned Active Learning for Heterogeneous Systems

Abstract Active learning is a subfield of machine learning that focuses on improving the data collection efficiency in expensive-to-evaluate systems. Active learning-applied surrogate modeling facilitates cost-efficient analysis of demanding engineering systems, while the existence of heterogeneity in underlying systems may adversely affect the performance. In this article, we propose the partitioned active learning that quantifies informativeness of new design points by circumventing heterogeneity in systems. The proposed method partitions the design space based on heterogeneous features and searches for the next design point with two systematic steps. The global searching scheme accelerates exploration by identifying the most uncertain subregion, and the local searching utilizes circumscribed information induced by the local Gaussian process (GP). We also propose Cholesky update-driven numerical remedies for our active learning to address the computational complexity challenge. The proposed method consistently outperforms existing active learning methods in three real-world cases with better prediction and computation time.

Active instance selection via parametric equation and instance overlap aware scheme

Active learning (AL) reduces the human labeling effort by learning a classifier with the small labeled data. This small labeled data is formed by those instances, which reduce generalization error the most. As AL reduces the labeling cost, but most of the pool based query strategies evaluate all unlabeled instances in each iteration of query instance selection, which makes them computationally extensive. Also, many times the query strategies lead to the selection of redundant/overlapped instances, which leads to no improvement in the generalization performance. In this work, an advanced query strategy under pool based scenario has been proposed. This query strategy utilizes parametric equation based query synthesis as the informative criterion and the instance overlap aware scheme as the representative criterion. This informative criterion identifies the input instances near the decision boundary, which speeds up the instance selection process and hence reduces the response time. On the set of identified instances, the representative criterion is used to avoid the overlapped instance selection, which improves the generalization performance. Also, a meta-learning based approach has been used to identify the values of the main parameters used in the formulated criteria. The comparison of the proposed approach with the existing baseline solutions on artificial as well as real-world datasets demonstrates that the proposed approach achieves a significant reduction in the query instance selection time, along with an enhancement in the generalization performance over the existing approaches.

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.

Student Perceptions of Active-Learning Strategies Utilized in a Drugs in Pregnancy Elective Delivered Synchronously Across Multiple Campuses

Active-learning activities can be successfully implemented into a course even when instructors are not in the same physical classroom as students. Technology can assist with synchronous distance education as well as facilitate active learning. This manuscript describes student perceptions of active learning activities in a re-designed drugs in pregnancy course that was delivered synchronously to two student campuses with the instructor at an off-campus site. Small and large group discussions, team-based learning activities, case scenario exercises, and a debate were conducted within the course. Padlet™ was the primary software technology incorporated into learning activities. Students reported positive perceptions of the various active learning activities and the technology used in the course. Students also stated they were beneficial for learning. In the post-course survey, students indicated that the course met their expectations overall. Students’ confidence levels for completing tasks related to the course materials significantly increased after completion of the course.

Jointly Informative and Manifold Structure Representative Sampling Based Active Learning for Remote Sensing Image Classification

Active learning (AL) methods that select unlabeled samples only querying by informative measures (i.e., uncertainty and/or diversity criteria) have been extensively investigated. However, these methods usually do not exploit the manifold structure of the unlabeled data from the geometrical point of view, a choice that might lead to a sample bias and consequently undesirable performances. To control and possibly overcome such drawbacks, this paper explores AL methods based on joint informative and manifold structure representative sampling (JI-MSRS). In JI-MSRS, a portion of the unlabeled samples that are added at each iteration is selected according to the informative measures, whereas another portion is selected according to their capability to represent the data cluster structure. Four popular manifold learning methods, namely, principle component analysis (PCA), linear discriminant analysis, kernel PCA, and neighborhood preserving embedding, are used to model the data structure. Then, Delaunay triangulation nets are used to build a discrete approximation of the geometrical structure of the unlabeled data cloud in a low-dimensional space. To show the effectiveness of this novel sampling strategy, results on three real multi-/hyperspectral data sets are presented, adding a thorough comparison with other state-of-the-art AL techniques. In comparison to conventional AL heuristics, the proposed techniques are able to obtain competitive or even better classification accuracy values.

Using Interactive Methods in Teaching Accounting

Abstract Active learning shifts the focus of instruction from what the teacher should teach or deliver to students, to what the students should be able to do with the course material. Interactive methods are modern ways of stimulating teaching, representing tools of learning which favor the interchange of ideas, experiences, and knowledge. The purpose of this paper is to present a series of active teaching methods for the accounting data as being an alternative in the process of improving the teaching and learning of the accounting process.

Patch-based active learning (PTAL) for spectral-spatial classification on hyperspectral data

Active learning (AL) has been shown to be a useful approach to improving the efficiency of the classification process for remote-sensing imagery. Current AL methods are essentially based on pixel-wise classification. In this paper, a new patch-based active learning (PTAL) framework is proposed for spectral-spatial classification on hyperspectral remote-sensing data. The method consists of two major steps. In the initialization stage, the original hyperspectral images are partitioned into overlapping patches. Then, for each patch, the spectral and spatial information as well as the label are extracted. A small set of patches is randomly selected from the data set for annotation, then a patch-based support vector machine (PTSVM) classifier is initially trained with these patches. In the second stage (close-loop stage of query and retraining), the trained PTSVM classifier is combined with one of three query methods, which are margin sampling (MS), entropy query-by-bagging (EQB), and multi-class level uncertainty (MCLU), and is subsequently employed to query the most informative samples from the candidate pool comprising the rest of the patches from the data set. The query selection cycle enables the PTSVM model to select the most informative queries for human annotation. Then, these informative queries are added to the training set. This process runs iteratively until a stopping criterion is met. Finally, the trained PTSVM is employed to patch classification. In order to compare this to pixel-based active learning (PXAL) models, the prediction label of a patch by PTSVM is transformed into a pixel-wise label of a pixel predictor to get the classification maps. Experimental results show better performance of the proposed PTAL methods on classification accuracy and computational time on three different hyperspectral data sets as compared with PXAL methods.

Comparison of the use of role-play in both large and small classes to teach the basic concepts of an immune response to undergraduate students (41.3)

Abstract Active learning exercises are touted as superior teaching strategies, yet little work has been done to truly evaluate their effectiveness in the science classroom. Role-play is considered an especially strong tool to model concepts and show relationships between entities. The immune response is a particularly difficult topic for students to grasp, due to the myriad of cellular and protein interactions involved. This study looks at the use of role-play as a tool to teach the immune response to undergraduates in both small classes, where all students participate, and large classes where only a subset of students act for their peers. Learning was measured using a standard pre- and post-test format of multiple choice questions, and student perceptions of learning were collected using an anonymous written survey. No significant differences were found in overall learning based upon class size. Role-play participation did not have a significant impact on learning compared to students who did not participate in large classes. No significant differences were seen in learning based upon student learning styles. A full analysis of the efficacy of role-play as a replacement for or an addition to lecture will be presented. This research gives insight into the efficacy of role-play as a learning tool for teaching basic concepts in immunology.

Active learning improves comprehension in large section immunology course (37.7)

Abstract Active learning eschews the traditional sage-on-stage model of education in favor of a collaborative learning environment where students share with faculty the responsibility for making learning happen. The approach is especially well suited for students in large lecture classes who might otherwise feel isolated, unengaged or reluctant to interact. At University of Minnesota, we use a variety of active learning techniques in our large section, upper level undergraduate immunology course. Active learning has allowed us to incorporate formative assessment as an ongoing aspect of our class, as well as to encourage active engagement of a wide base of students in discussions. We describe here an active learning module on the complement system and report on the relative efficacy of the approach in fostering student achievement. The complement system is a conceptually trying topic for the typical undergraduate student of immunology. Results of pre- and post-tests over three academic years indicate students in the active learning intervention score higher than those in the control group. However, despite superior knowledge gains, student satisfaction with the learning experience declines.