Active Learning Methods Research Articles

Active learning of the collision distance function for high-DOF multi-arm robot systems

Abstract Motion planning for high-DOF multi-arm systems operating in complex environments remains a challenging problem, with many motion planning algorithms requiring evaluation of the minimum collision distance and its derivative. Because of the computational complexity of calculating the collision distance, recent methods have attempted to leverage data-driven machine learning methods to learn the collision distance. Because of the significant training dataset requirements for high-DOF robots, existing kernel-based methods, which require $O(N^2)$ memory and computation resources, where $N$ denotes the number of dataset points, often perform poorly. This paper proposes a new active learning method for learning the collision distance function that overcomes the limitations of existing methods: (i) the size of the training dataset remains fixed, with the dataset containing more points near the collision boundary as learning proceeds, and (ii) calculating collision distances in the higher-dimensional link $SE(3)^n$ configuration space – here $n$ denotes the number of links – leads to more accurate and robust collision distance function learning. Performance evaluations with high-DOF multi-arm robot systems demonstrate the advantages of the proposed active learning-based strategy vis- $\grave{\text{a}}$ -vis existing learning-based methods.

A sequential sampling-based Bayesian numerical method for reliability-based design optimization

For efficiently solving the Reliability-Based Design Optimization (RBDO) problem with multi-modal, highly nonlinear and expensive-to-evaluate limit state functions (LSFs), a sequential sampling-based Bayesian active learning method is developed in this work. The penalty function method is embedded to transform the constrained optimization problem into a non-constrained one to reduce the model complexity. The proposed method for solving RBDO problems starts by training a Gaussian process (GP) model, in the augmented space of random and design variables. It is then based on an efficient sampling scheme for simulating the GP model, the adaptive Bayesian optimization (BO) and Bayesian reliability analysis (BRA) procedures are combined in a collaborative way for sequentially producing the joint training points. BO driven by expected improvement (EI) function is used for inferring the global optimum in the design space with global convergence, and the BRA equipped with U function is implemented for inferring the failure probabilities at the identified design points with the desired accuracy. The superiority of the proposed method is demonstrated with two numerical and two real-world engineering examples.

Pool-based active learning framework for concept prerequisite learning

Concept prerequisite relation refers to the learning order of concepts, which is useful in education. Concept prerequisite learning refers to using machine learning methods to infer prerequisite relation of a concept pair. The process of concept prerequisite learning requires large amounts of labeled data to train classifier. Usually, the labels of prerequisite relation are assigned by specialists. The specialist labelling method is costly. Thus, it is necessary to reduce labeling expense. An effective strategy is using active learning methods. In this paper, we propose a pool-based active learning framework for concept prerequisite learning named PACOL. It is a fact that concept u and concept v cannot be prerequisite of each other simultaneously. The idea of PACOL is to select the concept pair with the greatest deviation between the classifier’s prediction and the fact. Besides, PACOL can be used in two situations: when specialists assign three kinds of labels or two kinds of labels. In experiments, we constructed data sets for three subjects. Experimental results on both our constructed data sets and public data sets demonstrate that PACOL outperforms than existing active learning methods in all situations.

A project-based learning (PBL) on gas dynamic concepts using hydraulic analogy technique

This article suggests the hydraulic analogy technique as a cost-effective approach for flow visualization in fluid mechanics, particularly in dealing with challenges related to supersonic flow, such as shock waves and flow past wedges. The project-based learning approach is adopted to engage students in the study effectively. This active learning method allows students to actively participate in research activities, from setting project goals to designing and executing experiments. The article highlights the positive impact of project-based learning on students’ learning outcomes through internal and external evaluations. The study shows that project-based learning enhances technical competence and encourages a lifelong passion for learning. The cost-effectiveness of the hydraulic analogy method inspires neighbouring academic institutions to consider similar approaches, further promoting the use of practical and affordable solutions for flow visualization experiments.

Implementasi Pendidikan Karakter pada Sekolah Inklusif di SDN Cisarua Kota Sukabumi


 
 
 
 This research aims to describe the implementation of character education in inclusive schools at SDN Cisarua, Sukabumi City. The aspects observed in the implementation of character education include learning, example, strengthening, and habituation. This is qualitative research with a descriptive qualitative research design. The subjects of this research included school principals, class teachers, and students with special needs (ABK). The data collection techniques used in this research are interviews, observation, and documentation studies. This research data was analyzed using data reduction steps, data presentation, and drawing conclusions. Data validity checking technique using triangulation of techniques and sources the results of the research show that the principal’s policy and the role of the class teacher have been instilled in students in their class, especially those with special needs, through learning, example, strengthening, and habituation. In implementing learning, teachers emphasize instilling the character values of tolerance and caring. The teacher does this in learning by instilling concepts through explanations, discussing moral issues, stories, active learning, and collaborative methods. Apart from that, teachers also set examples in attitudes and actions, give awards, provide individual assistance, and familiarize students with mingling with friends with special needs, both in the classroom and outside the classroom.
 
 
 

Virtual Active Learning to Maximize Knowledge Acquisition in Nursing Students: A Comparative Study.

Nursing students need to acquire knowledge through active methods that promote critical thinking and decision making. The purpose of this study is to analyze whether there are differences in the acquisition of knowledge by nursing students between active face-to-face or virtual teaching methods. In this comparative study, nursing students enrolled in the psychology course were divided into two groups: a face-to-face group that received active teaching methods and a virtual group. The virtual group was exposed to the Effective Learning Strategy (ELS), which included seminars based on video content through the Virtual Campus and answering questions using the H5P tool. In addition, participants engaged in reflection tasks on the content. Covariate data were collected, and knowledge tests were administered to both groups before and after the course. After three months, subjects were re-evaluated with a final exam to assess content retention. A total of 280 students were randomized. No differences were found in students' scores at the end of the knowledge test or in their final grades in the subject. Having study habits (b = 0.12, p = 0.03) and social support from relevant people (b = 0.09; p = 0.03) were associated with better post-intervention scores, and inversely with social support from friends (b = -0.12, p < 0.01). Final grades were inversely associated with digital safety literacy (b = -0.101, p = 0.01). No factors were associated with the scores of each group separately. The ELS virtual active learning model is as effective as face-to-face active learning methods for teaching psychology to first-year nursing students. This study was not registered.

You only label once: A self‐adaptive clustering‐based method for source‐free active domain adaptation

AbstractWith the growing significance of data privacy protection, Source‐Free Domain Adaptation (SFDA) has gained attention as a research topic that aims to transfer knowledge from a labeled source domain to an unlabeled target domain without accessing source data. However, the absence of source data often leads to model collapse or restricts the performance improvements of SFDA methods, as there is insufficient true‐labeled knowledge for each category. To tackle this, Source‐Free Active Domain Adaptation (SFADA) has emerged as a new task that aims to improve SFDA by selecting a small set of informative target samples labeled by experts. Nevertheless, existing SFADA methods impose a significant burden on human labelers, requiring them to continuously label a substantial number of samples throughout the training period. In this paper, a novel approach is proposed to alleviate the labeling burden in SFADA by only necessitating the labeling of an extremely small number of samples on a one‐time basis. Moreover, considering the inherent sparsity of these selected samples in the target domain, a Self‐adaptive Clustering‐based Active Learning (SCAL) method is proposed that propagates the labels of selected samples to other datapoints within the same cluster. To further enhance the accuracy of SCAL, a self‐adaptive scale search method is devised that automatically determines the optimal clustering scale, using the entropy of the entire target dataset as a guiding criterion. The experimental evaluation presents compelling evidence of our method's supremacy. Specifically, it outstrips previous SFDA methods, delivering state‐of‐the‐art (SOTA) results on standard benchmarks. Remarkably, it accomplishes this with less than 0.5% annotation cost, in stark contrast to the approximate 5% required by earlier techniques. The approach thus not only sets new performance benchmarks but also offers a markedly more practical and cost‐effective solution for SFADA, making it an attractive choice for real‐world applications where labeling resources are limited.

Evaluating face-to-face and online flipped learning on performance and satisfaction in marketing and communication students

The research evaluates the impact of face-to-face and online flipped learning (OFL) on the academic performance in students of single-degree (SD) and double-degree (DD), compared to that achieved in the traditional methodology. A descriptive, quasi-experimental, cross-sectional, quantitative study was carried out with a sample of 223 university marketing and communication students. The study was conducted in three phases: before, during, and after the COVID-19 pandemic. Results align with previous studies, which find that flipped learning (FL) face-to-face yields higher performance than traditional methodology. Like other studies, the highest scores in asynchronous activities were found for DD students. However, this study reveals that those adopting FL, both online and face-to-face, demonstrated superior academic performance compared to SD students using the traditional method in practical assessments. Notably, DD students who used OFL methodology outperformed their SD peers who used the traditional method in the overall subject score and in the synchronous and asynchronous activities. Furthermore, although modality did not influence the ratings of FL methodology; DD students who experienced this methodology online or face-to-face reported more positively on their attitudes, perceptions, interactions, academic results, and perceived satisfaction than SD students. These findings contribute to understanding of how the adoption of diverse learning methods and modalities influence the performance, attitudes, perceptions, interactions, results, and satisfaction of SD and DD students in the fields of marketing and communication. In conclusion, university business education institutions can enhance student performance and satisfaction by expanding DDs offerings and integrating active learning methods.

Uncertainty-weighted prototype active learning in domain adaptive semantic segmentation

Generalizing segmentation networks that adapt to unseen target domains is practical to the real-world utility. Recent self-training-based approaches compute the pseudo labels of highly imbalanced target image that are typically biased to the majority classes and basically noisy. To avoid the model falling into an error-prone and suboptimal state, we propose an uncertainty-weighted prototype active learning method to deal with domain adaptation regarding the semantic segmentation task. For assisting active learning to select reliable samples, multiple prototypical anchors strategy is employed to build the target image distribution model and assign accurate pixel-level pseudo labels, which consists of two different stages. In the initial training stage, this model defines each class as a set of non-learnable prototypes, and performs online clustering to identify target instances for annotating in the prototype-based feature space. In the fine-tuning stage, entropy criterion is adopted to generate representative class prototypes set of target domain, then compute the uncertainty as the clustering weight of prototype-level features via a predictive entropy-based strategy. By mixing these latent representations of the target samples and developing feature alignment loss function, our method can achieve more accurate segmentation performance. Extensive experiments on two benchmarks, GTAV → Cityscapes and SYNTHIA → Cityscapes, demonstrate that the proposed model outperforms other state-of-the-art approaches, and ablation study verifies the effectiveness of each key component. We hope our work will open a new view for future research in the field of domain adaptation. The code will be released soon at https://github.com/zihaodong/UPAL.

AI-Enhanced Personalized Learning Practices in Higher Engineering Institutes

The main task of engaging the students with an interesting classroom environment is challenging in nature. The better way of teaching the students with effective learning pedagogy seems to be more effective. In this technological era, every student requires some technological involvement in their day-to-day process. So, with a broadcasted mindset, an active learning pedagogy has been initialized in this manuscript which gives more and more active methods of learning especially for engineering students. This research mainly focuses on a widely available learning process that is more effective in context with an easy understanding of the concepts. The AI – Enhanced Personalized learning methodology has been utilized in this manuscript. A deep understanding of the concept lies in the visual representation of the concept. In context to this, the simulation-based active learning pedagogy has been implemented for the Electrical Engineering graduates of the third year. The pedagogy related cource outcome (CO) has been framed. The effectiveness of the implemented method has been evaluated with various feedback responses and their scores in the end-semester with and without AI methodology help to view the performance metrics of the implemented AI – based learning learning pedagogy. Surely, this method yields good results in the performance of every student. Keywords – Active learning methodology, Teaching pedagogy, AI-based learning methodology, Program Outcome, Renewable energy sources.

«Զբոսաշրջության բիզնեսի պլանավորում և կազմակերպում» առարկայի դասավանդման մեթոդիկան ԲՊՀ «Զբոսաշրջություն» մագիստրոսական ծրագրում / Teaching Methodology of the Course "Tourism Business Planning and Organization" Within Master Program in "Tourism" at BSU

Business planning and organization courses are among the most popular in both formal and informal educational systems. This article aims to present a general framework for designing courses in business planning, development, and organization. It can serve as a methodological guide for creating these courses for both humanities and non-humanities disciplines. The course is designed to give a general insight into the economic and commercial activities in the field of tourism. It incorporates both passive and active learning methods, including lectures, individual/practical assignments, and project-based learning. The objective of the course is to equip students with fundamental knowledge and skills in entrepreneurial initiative, specifically tailored to the training of future tourism specialists.

Perceived Faculty Barriers to Using Active Learning Methods in the Higher Education Classroom

Perceived Faculty Barriers to Using Active Learning Methods in the Higher Education Classroom

СОЦИАЛЬНАЯ КОМПЕТЕНТНОСТЬ СПЕЦИАЛИСТОВ ЖЕЛЕЗНОДОРОЖНОГО ТРАНСПОРТА

Railway transport is the leader of the Russian transport market and one of the largest companies in the global transport sector. This powerful tool not only serves as a driving force for the development of many industries in the Russian Federation, but is also one of the criteria for assessing well-being and welfare of its citizens, which dictates the need for a detailed study of all aspects of the functioning of railway production. When considering the quality of the implementation of railway services, special attention must be paid to the internal quality system, namely, corporate culture, working conditions, quality of service, and a multi-level team management system. In modern rapidly changing socio-technological conditions of development of the transport industry, the efficiency and competitiveness of a railway transport specialist is determined not only by the quality of knowledge and skills in the structure of professional competencies, but also by a set of personal qualities that allow optimal ergonomic passage through the stages of socio-technological adaptation and further professionalization. A multifactorial analysis of “social competence” as a pedagogical category, from the standpoint of activity-based, personal, subjective, axiological, competence-based approaches, the study of regulatory documentation and conditions for the implementation of professional activities in transport, made it possible to determine the essence and structure of the social competence of railway specialists as an integrative quality of personality, the structure of which can be represented by a set of motivational, value, emotional, and communicative components. The choice of indicators (social initiative and activity, the ability to transmit generally accepted and intra-industry social values, empathy, tolerance, self-esteem, ability for effective communication, social mobility) in the structure of social competence is substantiated and the levels of their formation are determined. Situational and functional approaches to identifying the content of the main stages of socialization in the professional training of railway transport specialists made it possible to specify the psychological and pedagogical conditions for the targeted formation of indicators of social competence: incessancy and continuity in the development of a block of socially oriented disciplines; practice-oriented educational process; active learning methods; design technologies; formation of an active subject position of students

Hybrid retrieval of grass biophysical variables based-on radiative transfer, active learning and regression methods using Sentinel-2 data in Marakele National Park

Biophysical variables such as leaf area index (LAI) and leaf chlorophyll content (LCC) are cited as essential biodiversity variables. A comprehensive comparison and integration of retrieval methods is needed for the estimation of biophysical variables such as LAI and LCC over a multispecies grass canopy. This study tested an assortment of five potentially robust, nonparametric regression methods (NPRMs) for inversion of radiative transfer model (RTM) to retrieve grass LAI and LCC in the Marakele National Park (MNP) of South Africa. The NPRMs used were, namely (i) Partial least squares regression (PLSR), (ii) Principle components regression (PCR), (iii) Kernel ridge regression (KRR), (iv) Random forest regression (RFR), and (v) K-nearest neighbours regression (KNNR). Furthermore, the study attempted to constrain the inversion process by using active learning (AL) techniques which ensured the selection of informative samples from a large pool of RTM simulations. Results show the most accurate grass LAI and LCC retrievals had lower relative root mean squared errors (RRMSEs) of 39.87% and 16.58% respectively. These findings have significant implications for the development of transferable rangeland monitoring systems in protected mountainous regions.

Active learning of alchemical adsorption simulations; towards a universal adsorption model.

Adsorption is a fundamental process studied in materials science and engineering because it plays a critical role in various applications, including gas storage and separation. Understanding and predicting gas adsorption within porous materials demands comprehensive computational simulations that are often resource intensive, limiting the identification of promising materials. Active learning (AL) methods offer an effective strategy to reduce the computational burden by selectively acquiring critical data for model training. Metal-organic frameworks (MOFs) exhibit immense potential across various adsorption applications due to their porous structure and their modular nature, leading to diverse pore sizes and chemistry that serve as an ideal platform to develop adsorption models. Here, we demonstrate the efficacy of AL in predicting gas adsorption within MOFs using "alchemical" molecules and their interactions as surrogates for real molecules. We first applied AL separately to each MOF, reducing the training dataset size by 57.5% while retaining predictive accuracy. Subsequently, we amalgamated the refined datasets across 1800 MOFs to train a multilayer perceptron (MLP) model, successfully predicting adsorption of real molecules. Furthermore, by integrating MOF features into the AL framework using principal component analysis (PCA), we navigated MOF space effectively, achieving high predictive accuracy with only a subset of MOFs. Our results highlight AL's efficiency in reducing dataset size, enhancing model performance, and offering insights into adsorption phenomenon in large datasets of MOFs. This study underscores AL's crucial role in advancing computational material science and developing more accurate and less data intensive models for gas adsorption in porous materials.

Dual-Reference Source-Free Active Domain Adaptation for Nasopharyngeal Carcinoma Tumor Segmentation across Multiple Hospitals.

Nasopharyngeal carcinoma (NPC) is a prevalent and clinically significant malignancy that predominantly impacts the head and neck area. Precise delineation of the Gross Tumor Volume (GTV) plays a pivotal role in ensuring effective radiotherapy for NPC. Despite recent methods that have achieved promising results on GTV segmentation, they are still limited by lacking carefully-annotated data and hard-to-access data from multiple hospitals in clinical practice. Although some unsupervised domain adaptation (UDA) has been proposed to alleviate this problem, unconditionally mapping the distribution distorts the underlying structural information, leading to inferior performance. To address this challenge, we devise a novel Sourece-Free Active Domain Adaptation framework to facilitate domain adaptation for the GTV segmentation task. Specifically, we design a dual reference strategy to select domain-invariant and domain-specific representative samples from a specific target domain for annotation and model fine-tuning without relying on source-domain data. Our approach not only ensures data privacy but also reduces the workload for oncologists as it just requires annotating a few representative samples from the target domain and does not need to access the source data. We collect a large-scale clinical dataset comprising 1057 NPC patients from five hospitals to validate our approach. Experimental results show that our method outperforms the previous active learning (e.g., AADA and MHPL) and UDA (e.g., Tent and CPR) methods, and achieves comparable results to the fully supervised upper bound, even with few annotations, highlighting the significant medical utility of our approach. In addition, there is no public dataset about multi-center NPC segmentation, we will release code and dataset for future research (Git).

Conformal active learning-aided screening of ligand-protected Cu-nanoclusters for CO2 reduction reactions

In this study, we propose a conformal active learning (CAL) method to screen ligand-protected atomically precise Cu-nanoclusters for the CO2 reduction reaction.

CASL: Capturing Activity Semantics Through Location Information for Enhanced Activity Recognition.

Using portable tools to monitor and identify daily activities has increasingly become a focus of digital healthcare, especially for elderly care. One of the difficulties in this area is the excessive reliance on labeled activity data for corresponding recognition modeling. Labeled activity data is expensive to collect. To address this challenge, we propose an effective and robust semi-supervised active learning method, which combines the mainstream semi-supervised learning method with expert collaboration. Our method takes a user's trajectory as the only input. In addition, this novel method uses expert collaboration to judge the valuable samples further to enhance its performance. Our method relies on very few semantic activities, outperforms all baseline activity recognition methods, and is close to the performance of supervised learning methods. On the adlnormal dataset with 200 semantic activities data, our work achieved an accuracy of 89.07%, and supervised learning has 91.77%. Our ablation study validated the components in our method using a query strategy and a data fusion approach.

Methodological Innovation Path of Ideological and Political Education Content in Colleges and Universities in the Age of Artificial Intelligence

Abstract Under the background of the continuous development of modern curriculum as well as the continuous innovation of educational methods, the content and methods of ideological and political education in colleges and universities are also continuously innovated. This paper extracts the features of massive Civic and political knowledge points and combines them with active learning methods to construct the relationship between the prerequisites of Civic and political knowledge points, and then introduces the cognitive diagnostic model to design the cognitive diagnostic model based on the requirements of Civic and political knowledge points. Accurate and rich auxiliary information, such as students’ mental state, is provided for Civics teaching, and a Civics education model is designed based on this model. Finally, the teaching model is applied to actual teaching to verify its effectiveness. The results show that after completing the teaching practice, the students in the experimental class scored in “political identity,” “professionalism,” “public participation,” and “sound personality.” Sound personality” are all above 22. Compared with the control class, there are significant differences in all dimensions except “sound personality” (P&lt;0.05). A cognitive diagnosis-based civic education model can significantly improve students’ level of civic education.

An Active Learning Method for Solving Competitive Multi-Agent Decision-Making and Control Problems

An Active Learning Method for Solving Competitive Multi-Agent Decision-Making and Control Problems