Collaborative Learning Framework Research Articles

Key protected classification for collaborative learning

Large-scale datasets play a fundamental role in training deep learning models. However, dataset collection is difficult in domains that involve sensitive information. Collaborative learning techniques provide a privacy-preserving solution, by enabling training over a number of private datasets that are not shared by their owners. However, recently, it has been shown that the existing collaborative learning frameworks are vulnerable to an active adversary that runs a generative adversarial network (GAN) attack. In this work, we propose a novel classification model that is resilient against such attacks by design. More specifically, we introduce a key-based classification model and a principled training scheme that protects class scores by using class-specific private keys, which effectively hide the information necessary for a GAN attack. We additionally show how to utilize high dimensional keys to improve the robustness against attacks without increasing the model complexity. Our detailed experiments demonstrate the effectiveness of the proposed technique. Source code will be made available at https://github.com/mbsariyildiz/key-protected-classification.

Collaborative learning based on centroid-distance-vector for wearable devices

In typical application scenarios of Internet-of-Things (IoT), human activity recognition capability of a wearable device is usually limited to the body segment on which the device is worn. We believe that an efficient approach where wearable devices collaboratively learn can lead to better performance of activity recognition. In this paper, we first propose the concept of centroid distance vector to fully exploit and express the hidden knowledge of available data. Then, we introduce our centroid-distance-vector-based collaborative learning framework which consists of two key steps: data expansion with centroid distance vector and automatic model construction. In order to achieve automatic model construction, we further propose two collaborative learning methods: a centroid-distance-vector-based label propagation method (CD-LP) and a centroid-distance-vector-based knowledge distillation method (CD-KD). Experiments have shown that by using our proposed algorithms, the human activity recognition accuracy can achieve an average improvement of 5.24% compared to the state-of-the-art machine learning methods, and 5.88 % compared to the neural network with two fully-connected hidden layers.

Collaborative learning of lightweight convolutional neural network and deep clustering for hyperspectral image semi-supervised classification with limited training samples

Deep learning provides excellent potentials for hyperspectral images (HSIs) classification, but it is infamous for requiring large amount of labeled samples while the collection of high-quality labels for HSIs is extremely expensive and time-consuming. Furthermore, when the limited training samples are available, deep learning methods may suffer from over-fitting. In this work, we propose a novel collaborative learning framework for semi-supervised HSI classification with joint deep convolutional neural networks and deep clustering. Specifically, a lightweight 3D convolutional neural network (CNN) with much less parameters compared with classical 3D CNNs is designed for deep discriminative feature learning and classification. Then a deep clustering method, that is approximate rank-order clustering (AROC) algorithm, is applied to cluster deep features to generate pseudo labels for abundant unlabeled samples. Finally, we fine-tune the lightweight 3D CNN by minimizing a dual-loss (softmax loss and center loss) using both true and pseudo labels. Experimental results on three challenging HSI datasets demonstrate that the proposed method can achieve better performance than other state-of-the-art deep learning based methods and traditional HSI classification methods methods.

Multi-indices quantification of optic nerve head in fundus image via multitask collaborative learning.

Multi-indices quantification of optic nerve head (ONH), measuring ONH appearance with multiple types of indices simultaneously from fundus images, is the most clinically significant tasks for accurate ONH assessment and ophthalmic disease diagnosis. However, no attempt has been reported due to its challenges of the large variation of fundus appearance across patients, heavy overlap and extremely weak contrast between optic nerve head areas. In this paper, we propose a multitask collaborative learning framework (MCL-Net) for multi-indices ONH quantification. The proposed MCL-Net, a two-branch neural network, first obtains expressive shared and task-specific representations with the backbone network and its two branches; then models the feature exchanges and aggregations between two branches with a well-designed feature interaction module (FIM) to promote each other collaboratively. After that, it estimates multiple types of ONH indices under a multitask ensemble module (MEM) that is capable of learning aggregation of multiple outputs automatically. Therefore, the proposed MCL-Net is consisted of the feature representation, inter-task feature interaction, dual-branch task-specific prediction, and multitask quantification ensemble, which establish an effective framework which takes full advantages of segmentation and estimation tasks for multi-indices ONH quantification. Rather than the low-level feature sharing and individual prediction, the proposed MCL-Net collaboratively learns an optimal combination of shared and task-specific representation, as well as the aggregated prediction, therefore leads to accurate quantification of ONH with multiple types of indices. Experimental results on the dataset of 650 fundus images show that MCL-Net successfully delivers accurate quantification of all the three types of ONH indices, with average mean absolute error of 0.98 ± 0.20, 0.97 ± 0.16, 1.19 ± 0.18, as well as average correlation coefficient of 0.699, 0.708 and 0.691, for diameters, whole areas and regional areas, respectively. In addition, the experiments demonstrate that quantitative indices obtained by our method provide more effective glaucoma diagnosis with AUC of 0.8698. This endows our proposed MCL-Net a great potential in clinical assessment from focal to global for ophthalmic disease diagnosis.

Intelligent Resource Scheduling for 5G Radio Access Network Slicing

It is widely acknowledged that network slicing can tackle the diverse use cases and connectivity services of the forthcoming next-generation mobile networks (5G). Resource scheduling is of vital importance for improving resource-multiplexing gain among slices while meeting specific service requirements for radio access network (RAN) slicing. Unfortunately, due to the performance isolation, diversified service requirements, and network dynamics (including user mobility and channel states), resource scheduling in RAN slicing is very challenging. In this paper, we propose an intelligent resource scheduling strategy (iRSS) for 5G RAN slicing. The main idea of an iRSS is to exploit a collaborative learning framework that consists of deep learning (DL) in conjunction with reinforcement learning (RL). Specifically, DL is used to perform large time-scale resource allocation, whereas RL is used to perform online resource scheduling for tackling small time-scale network dynamics, including inaccurate prediction and unexpected network states. Depending on the amount of available historical traffic data, an iRSS can flexibly adjust the significance between the prediction and online decision modules for assisting RAN in making resource scheduling decisions. Numerical results show that the convergence of an iRSS satisfies online resource scheduling requirements and can significantly improve resource utilization while guaranteeing performance isolation between slices, compared with other benchmark algorithms.

Deep Group-wise Fully Convolutional Network for Co-saliency Detection with Graph Propagation.

A key problem in co-saliency detection is how to effectively model the interactive relationship of a whole image group and the individual perspective of each image in a united data-driven manner. In this paper, we propose a group-wise deep co-saliency detection approach to address the co-saliency object discovery problem based on the fully convolutional network (FCN). The proposed approach captures the group-wise interaction information for group images by learning a semantics-aware image representation based on a convolutional neural network, which adaptively learns the group-wise features for co-saliency detection. Furthermore, the proposed approach discovers the collaborative and interactive relationships between group-wise feature representation and single image individual feature representation, and model this in a collaborative learning framework. Then, we set up a unified deep learning scheme to jointly optimize the process of group-wise feature representation learning and the collaborative learning, leading to more reliable and robust co-saliency detection results. Finally, we present a graph Laplacian regularized nonlinear regression model for saliency refinement. Experimental results demonstrate the effectiveness of our approach in comparison with the state-of-the-art approaches.

The design of an empirical cross-boundary collaborative open learning framework

This paper reports on the design and development of an empirical cross-boundary, collaborative open learning framework for cross-institutional academic development. The framework is one of the key outputs of a phenomenographic study into the lived experience of open learners in two open cross-institutional courses. Data was collected through individual interviews from 22 study participants from two courses that made up a collective case study. These courses were offered by UK higher education institutions in collaboration with informal partners nationally and internationally and were selected as these had different collaborative learning features built-in that could be explored. Course participants in these two courses include academic staff who teach or support learning and further open learners. The empirical design framework is proposed to support the development and implementation of cross-boundary collaborative open learning approaches within cross-institutional academic development and may also be useful in further learning and teaching settings in higher education.

Leveraging Prior-Knowledge for Weakly Supervised Object Detection Under a Collaborative Self-Paced Curriculum Learning Framework

Weakly supervised object detection is an interesting yet challenging research topic in computer vision community, which aims at learning object models to localize and detect the corresponding objects of interest only under the supervision of image-level annotation. For addressing this problem, this paper establishes a novel weakly supervised learning framework to leverage both the instance-level prior-knowledge and the image-level prior-knowledge based on a novel collaborative self-paced curriculum learning (C-SPCL) regime. Under the weak supervision, C-SPCL can leverage helpful prior-knowledge throughout the whole learning process and collaborate the instance-level confidence inference with the image-level confidence inference in a robust way. Comprehensive experiments on benchmark datasets demonstrate the superior capacity of the proposed C-SPCL regime and the proposed whole framework as compared with state-of-the-art methods along this research line.

Cross-Domain Collaborative Learning via Discriminative Nonparametric Bayesian Model

Cross-domain data analysis has been becoming more and more important, and can be effectively adopted for many applications. However, it is difficult to propose a unified cross-domain collaborative learning framework for cross-domain analysis in social multimedia, because cross-domain data have multidomain, multimodal, sparse, and supervised properties. In this paper, we propose a generic cross-domain collaborative learning (CDCL) framework via a discriminative nonparametric Bayesian dictionary learning model for cross-domain data analysis. Compared with existing cross-domain learning methods, our proposed model mainly has four advantages: First, to address the domain discrepancy, we utilize the shared domain priors among multiple domains to make them share a common feature space. Second, to exploit the multimodal property, we use the shared modality priors to model the relationship between different modalities. Third, to deal with the sparse property of media data in one domain, our goal is to learn a shared dictionary to bridge different domains and complement each other. Finally, to make use of the supervised property, we exploit class label information to learn the shared discriminative dictionary, and utilize a latent probability vector to select different dictionary elements for representation of each class. Therefore, the proposed model can investigate the superiorities of different sources to supplement and improve each other effectively. In experiments, we have evaluated our model for two important applications including cross-platform event recognition and cross-network video recommendation. The experimental results have showed the effectiveness of our CDCL model for cross-domain analysis.

Web 2.0-Based Collaborative Learning Framework Promoting Lifelong Learning:Developing Sustainability Competencies

This paper aims to present a Web 2.0-based collaborative Learning framework based on the social constructivist theory promoting lifelong learning. The Web 2.0 provides conditions of collaborative learning, multiple modes and methods of learning and multiple perspective to achieve learning goals. This practical framework under the guidelines of design science research methodology allows learners to actively engage in the learning process, construct own knowledge and use to solve real-world problems collaboratively. Experts reviewed and confirmed the validity of the framework. Three successive iterations, working with 42 students in formal learning and Facebook members in informal learning were conducted to demonstrate and validate how learners started developing critical thinking, self-regulation and use of knowledge. The framework contributes to the practice of lifelong learning with emphasis on developing sustainability competencies that enable learners to gain more skills in applying different problem-solving frameworks to complex daily life problems.

An Innovative Web 2.0-Based Collaborative Learning Framework: Activities for The Initial Development of Social Negotiation Competencies

An Innovative Web 2.0-Based Collaborative Learning Framework: Activities for The Initial Development of Social Negotiation Competencies

Analysis of knowledge construction during group space activities in a flipped learning course

AbstractFlipped learning is a pedagogical approach that uses technology to deliver instructional content outside of class (individual space) and class time to engage in collaborative activities (group space). In a flipped learning course, the traditional classroom paradigm shifts and teachers become more facilitator than lecturer. Research on flipped learning is limited, in that studies are mostly conducted in postsecondary classrooms and focus on engagement and performance rather than the learning that occurs during collaboration. My study was designed to investigate group learning or knowledge construction during collaborative activities. Participants were middle school students from Hawaii. To identify group learning, I targeted a specific activity between three students in the group space. Using a computer‐supported collaborative learning framework, I recorded and analysed the group's verbal communication as they worked together. Moments of knowledge construction and interactions leading to those moments were analysed. Content analysis and lag sequential analysis revealed significant strategies used by students to construct knowledge. Some recommendations for practitioners include designing a need for teacher facilitation, incorporating questions that promote student discussion, and requiring students to reflect on their understanding. Future studies should include different K‐12 environments and focus on knowledge construction transfer between the individual and group learning spaces.

A Unified Collaborative Multikernel Fuzzy Clustering for Multiview Data

Clustering is increasingly important for multiview data analytics and current algorithms are either based on the collaborative learning of local partitions or directly derived global clustering from multikernel learning. In this paper, we innovate a clustering model that unifies the local partitions and global clustering in a collaborative learning framework. We first construct a common multikernel space from a set of basis kernels to better reflect clustering information of each individual view. Then, considering that joint local partitions would conform to the global clustering, we fuse the local partitions and global clustering guidance as a single objective function in accordance with fuzzy clustering form. The collaborative learning strategy enables the mutual and interactive clustering from local partitions and global clustering. The validation was performed over two synthetic and four public databases and the clustering accuracy was measured by normalized mutual information and rand index. The experimental results demonstrated that the proposed algorithm outperformed the related state-of-the-art algorithms in comparison, which included multitask, multikernel, and multiview clustering approaches.

SQL Collaborative Learning Framework Based on SOA

The research is focused on designing collaborative learning-oriented framework fulfilment service in teaching SQL Oracle 10g. Framework built a foundation of academic fulfilment service performed by a layer of the working unit in collaboration with Program Studi Manajemen Informatika. In the design phase defined what form of collaboration models and information technology proposed for Program Studi Manajemen Informatika by using a framework of collaboration inspired by the stages of modelling a Service Oriented Architecture (SOA). Stages begin with analyzing subsystems, this activity is used to determine subsystem involved and reliance as well as workflow between the subsystems. After the service can be identified, the second phase is designing the component specifications, which details the components that are implemented in the service to include the data, rules, services, profiles can be configured, and variations. The third stage is to allocate service, set the service to the subsystems that have been identified, and its components. Implementation framework contributes to the teaching guides and application architecture that can be used as a landing realize an increase in service by applying information technology.

A Collaborative Learning Framework for Estimating Many Individualized Regression Models in a Heterogeneous Population

Mixed-effect models (MEMs) have been found very useful for modeling complex dataset where many similar individualized regression models should be estimated. Like many statistical models, the success of these models builds on the assumption that a central tendency can effectively establish the population-level characteristics and covariates are sufficient to characterize the individual variation as derivation from the center. In many real-world problems, however, the dataset is collected from a rather heterogeneous population, where each individual has a distinct model. To fill in this gap, we propose a collaborative learning framework that provides a generic methodology for estimating a heterogeneous population of individualized regression models by exploiting the idea of “canonical models” and model regularization. By using a set of canonical models to represent the heterogeneous population characteristics, the canonical models span the modeling space for the individuals’ models, e.g., although each individual model is distinct, its model parameter vector can be represented by the parameter vectors of the canonical models. Theoretical analysis is also conducted to reveal a connection between the proposed method and the MEMs. Both simulation studies and applications on Alzheimer's disease and degradation modeling of turbofan engines demonstrate the efficacy of the proposed method.

The Game Making Framework for Collaborative Learning

Collaborative learning is a study group process that each member contributes their idea, information, experience, skill, and ability they have so it can be used on learning activity and improve members’ knowledge[1]. Digital game can become one of collaborative learning media that is not just for entertainment but also can be used as a unique and effective interactive learning media[2]. In conventional game making for collaborative learning, the research proves that it can improve excitement and contribution in learning activities[3]. But in game making, there are many technical issues for non-computer science students[3]. Based on many issues, designing framework is chosen to become a solution to help people in developing game as collaborative learning media by themselves. Because framework is a basic form of system [4], consist of various command, function, and benefit for developer in application development [5][6], so game making for collaborative learning is conducted again with support from framework designed by researcher. In order to know the effect from framework support in game making as collaborative learning media, this study is based on two subjects of non-computer science students and computer science students.

The Use of Social Media in E-Learning: A Metasynthesis

<p class="3">The adoption of social media in e-learning signals the end of distance education as we know it in higher education. However, it appears to have very little impact on the way in which open and distance learning (ODL) institutions are functioning. Earlier research suggests that a significant part of the explanation for the slow uptake of social media in e-learning lies outside of conventional factors attributed to distance learning reforms.</p><p class="3">This research used the conceptual framework for online collaborative learning (OCL)<em> </em>in higher education. Social media such as blogs, wikis, Skype or Google Hangout, Facebook; and even mobile apps, such as WhatsApp; could facilitate deep learning and the creation of knowledge in e-learning at higher educational institutions.</p><p class="3">This metasynthesis is an interpretative integration of peer-reviewed qualitative research findings on social media in e-learning. It includes a synthesis of data, research methods, and theories used to investigate social media in e-learning. Seven themes emerged from the data which have been recrafted into a framework for social media in e-learning as the final product. The proposed framework could be useful to instructional designers and academics who are interested in using modern learning theories and want to adopt social media in e-learning in higher education as a deep learning strategy.</p>

A Framework for Collaborative and Convenient Learning on Cloud Computing Platforms

The depth of learning resides in collaborative work with more engagement and fun. Technology can enhance collaboration with a higher level of convenience and cloud computing can facilitate this in a cost effective and scalable manner. However, to deploy a successful online learning environment, elementary components of learning pedagogy must be embedded in the technology framework. A cloud computing technology based learning platform built on the proven learning pedagogies will be most successful attempt to facilitate collaborative and convenient learning. Such platform will support the learners to connect and coordinate the online resources in a more efficient, effective and practical way. This paper proposes a new theoretical framework for collaborative and convenient learning on cloud computing technology by reviewing various learning theories and integrating them with the technology framework. Architectural details of the framework have been presented with suitable examples for a better understanding. The work has a lot of significance for deploying cloud computing based collaborative learning solutions in the educational institutions.

Rationale and methodology of a collaborative learning project in congenital cardiac care

Collaborative learning is a technique through which individuals or teams learn together by capitalizing on one another's knowledge, skills, resources, experience, and ideas. Clinicians providing congenital cardiac care may benefit from collaborative learning given the complexity of the patient population and team approach to patient care. Industrial system engineers first performed broad-based time-motion and process analyses of congenital cardiac care programs at 5 Pediatric Heart Network core centers. Rotating multidisciplinary team site visits to each center were completed to facilitate deep learning and information exchange. Through monthly conference calls and an in-person meeting, we determined that duration of mechanical ventilation following infant cardiac surgery was one key variation that could impact a number of clinical outcomes. This was underscored by one participating center's practice of early extubation in the majority of its patients. A consensus clinical practice guideline using collaborative learning was developed and implemented by multidisciplinary teams from the same 5 centers. The 1-year prospective initiative was completed in May 2015, and data analysis is under way. Collaborative learning that uses multidisciplinary team site visits and information sharing allows for rapid structured fact-finding and dissemination of expertise among institutions. System modeling and machine learning approaches objectively identify and prioritize focused areas for guideline development. The collaborative learning framework can potentially be applied to other components of congenital cardiac care and provide a complement to randomized clinical trials as a method to rapidly inform and improve the care of children with congenital heart disease.

From learners’ concept maps of their similar or complementary prior knowledge to collaborative concept map: Dual eye-tracking and concept map analyses

The present paper describes the MUPEMURE (Multiple Perspectives on Multiple Representations; Bodemer, Author, Kapur, Rummel, & Weinberger, 2011) model as a conceptual framework for collaborative learning with multiple external representations. Within this framework, a study was conducted to examine how learners working in dyads translated between self-generated concept maps of their own and partner's prior knowledge to create a collaborative concept map. Before individual and collaborative concept mapping (CCM) sessions, prior information on the learning topic was distributed between dyad members in such a way that they had either similar (“same/shared prior knowledge” or SK condition) or different but complementary knowledge (“different/unshared prior knowledge” or DK condition). The dual eye-tracking method was used in this study, and eye movement data was examined in two time periods, the first and second half of the CCM session. Eye movement analysis was complemented with the analysis of concept map measures, such as achievement, individual-to-group transfer and group creativity. In the first half of the CCM session, eye movement behavior across the three maps was interpreted as reflecting initial externalization and divergent processes through which learners mutually compare their knowledge and engage in a process of modeling each other. The second half of the CCM session was identified as a convergent phase characterized by a decrease in eye movement transitions between maps, and a convergence of attention on the collaborative map. The results also showed that in the first half of the collaboration, learners of the DK condition transitioned more between their own and partner maps, and therefore had more difficulty coordinating their respective prior knowledge compared to learners of the SK condition. Participants with different/unshared prior knowledge also focused their attention mainly in their own map and explored less their partner map in the first half of the collaboration. This suggests that they worked in a less collaborative way compared to those with same/shared prior knowledge during the divergent phase.