Joint Learning Research Articles

Joint learning of structural and textual information on propagation network by graph attention networks for rumor detection

Joint learning of structural and textual information on propagation network by graph attention networks for rumor detection

A machine reading comprehension framework for recognizing emotion cause in conversations

Recognizing Emotion Cause in Conversations (RECC) is a key issue in modeling human cognitive processes, involving Conversational Causal Emotion Entailment task (C2E2) and Conversational Causal Span Extraction task (C2SE). Previous emotion cause extraction research has been concentrated at the clause level, detecting if the cause is in the text, not describing the underlying causes in texts well. In order to address this issue, we suggest a novel approach that can recognize emotion cause spans. These spans can represent or imply the causes for controlling emotions. In this paper, we use a Machine Reading Comprehension framework to Recognize the Emotion Cause in Conversations (MRC-RECC), at both the span level and clause level simultaneously. Specifically, we use two types of queries to build the associations between the two different subtasks: emotion causal entailment task and emotion causal span extraction task. Our framework can recognize emotion cause more effectively by using joint learning to make these two tasks complement each other. Experiments demonstrate that our MRC-RECC provides state-of-the-art performances, which can reason more emotion causes in conversation texts. The code can be found at https://github.com/Guangzidetiaoyue/MRC-RECCON.

Vision-Based Multiscale Construction Object Detection under Limited Supervision

Contemporary multiscale construction object detection algorithms rely predominantly on fully-supervised deep learning, requiring arduous and time-consuming labeling process. This paper presents a novel semisupervised multiscale construction objects detection (SS-MCOD) by harnessing nearly infinite unlabeled images along with limited labels, achieving more accurate and robust detection results. SS-MCOD uses a deformable convolutional network (DCN)-based teacher-student joint learning framework. DCN uses deformable advantages to extract and fuse multiscale construction object features. The teacher module generates pseudolabels for construction objects in unlabeled images, while the student module learns the location and classification of construction objects in both labeled images and unlabeled images with pseudolabels. Experimental validation using commonly used construction datasets demonstrates the accuracy and generalization performance of SS-MCOD. This research can provide insights for other detection tasks with limited labels in the construction domain.

Shape embedding and retrieval in multi-flow deformation

We propose a unified 3D flow framework for joint learning of shape embedding and deformation for different categories. Our goal is to recover shapes from imperfect point clouds by fitting the best shape template in a shape repository after deformation. Accordingly, we learn a shape embedding for template retrieval and a flow-based network for robust deformation. We note that the deformation flow can be quite different for different shape categories. Therefore, we introduce a novel multi-hub module to learn multiple modes of deformation to incorporate such variation, providing a network which can handle a wide range of objects from different categories. The shape embedding is designed to retrieve the best-fit template as the nearest neighbor in a latent space. We replace the standard fully connected layer with a tiny structure in the embedding that significantly reduces network complexity and further improves deformation quality. Experiments show the superiority of our method to existing state-of-the-art methods via qualitative and quantitative comparisons. Finally, our method provides efficient and flexible deformation that can further be used for novel shape design.

Joint learning of multi-level dynamic brain networks for autism spectrum disorder diagnosis

Graph convolutional networks (GCNs), with their powerful ability to model non-Euclidean graph data, have shown advantages in learning representations of brain networks. However, considering the complexity, multilayeredness, and spatio-temporal dynamics of brain activities, we have identified two limitations in current GCN-based research on brain networks: 1) Most studies have focused on unidirectional information transmission across brain network levels, neglecting joint learning or bidirectional information exchange among networks. 2) Most of the existing models determine node neighborhoods by thresholding or simply binarizing the brain network, which leads to the loss of edge weight information and weakens the model's sensitivity to important information in the brain network. To address the above issues, we propose a multi-level dynamic brain network joint learning architecture based on GCN for autism spectrum disorder (ASD) diagnosis. Specifically, firstly, constructing different-level dynamic brain networks. Then, utilizing joint learning based on GCN for interactive information exchange among these multi-level brain networks. Finally, designing an edge self-attention mechanism to assign different edge weights to inter-node connections, which allows us to pick out the crucial features for ASD diagnosis. Our proposed method achieves an accuracy of 81.5 %. The results demonstrate that our method enables bidirectional transfer of high-order and low-order information, facilitating information complementarity between different levels of brain networks. Additionally, the use of edge weights enhances the representation capability of ASD-related features.

The Influence of the Learning Together Type Cooperative Learning Model Assisted by Smart Closet Media on Students' Social Sciences Learning Outcomes

Smart cupboard media is one of the simple media, which contains knowledge in it which is able to teach students to dare to express opinions or knowledge that they have gained so that it can grow students' ideas and insights, which makes students more focused in learning and speed in thinking so that they can improve results. student learning. The problem in this research is the low value of students' social studies learning outcomes. This research aims to determine (1) the application of the cooperative learning model of joint learning type assisted by smart cupboard media on the social studies learning outcomes of class VIII students at SMPN 1 Bandar Baru. (2) find out how much influence the learning outcomes have before and after using the cooperative learning model of the joint learning type assisted by smart cupboard media on the social studies learning outcomes of class VIII students at SMPN 1 Bandar Baru. This research uses a quantitative approach with a pre-experimental design research type, in the form of a one group pretest-posttest design where there is only one group. The population in this study was all 6 class VIII students. The sample in this study was 32 students in class VIII.2 who were given treatment in the form of implementing a cooperative learning model using the smart cupboard media. The sampling technique uses a purposive sampling technique. The data collection techniques used are tests and documentation. The data analysis technique uses the normality test, homogeneity test, and t-test for the difference between two means. The results of this research were obtained from the scores before using the cooperative learning model with the help of smart cupboard media and the posttest with an instrument in the form of 15 multiple choice questions. Based on hypothesis testing using the t test (paired t test), the tcount = 8,389 and the ttable value at the α = 0.05 level with df = 30 is 2,042, meaning tcount ttable or 8,389 2,042 at the Original Article Page 2 of 10 Rahmatillah et al. http://doi.org/10.24815/jimps.v9i1.xxxx significance level α = 5% (0 .05) then Ho is rejected and Ha is accepted or there is a real (significant) influence of the variable on the variable of the cooperative learning model type of joint learning assisted by smart cupboard media on learning outcomes in class VIII-2. Based on this description, it proves that the use of the cooperative learning model of the joint learning type assisted by smart cupboard media has a very positive effect on student learning outcomes in social studies subjects in class VIII.2 at SMP Negeri 1 Bandar Baru.

Causal Probabilistic Spatio-temporal Fusion Transformers in Two-sided Ride-Hailing Markets

In this work, we address the complex problem of multi-objective time series forecasting with a focus on predicting interdependent targets such as supply and demand in ride-hailing services. Traditional machine learning techniques approach the targets independently, while deep learning strategies may use joint learning with shared representations, both neglecting inter-target causal relationships and potentially compromising the models’ generalization capabilities. Our novel CausalTrans model introduces a framework to define and leverage the temporal causal interplay between supply and demand, incorporating both temporal and spatial causality into the forecasting process. Additionally, we enhance computational efficiency by introducing an innovative fast attention mechanism that reduces the time complexity from quadratic to linear without sacrificing performance. Our comprehensive experiments show that CausalTrans significantly surpasses contemporary forecasting methods, achieving up to a 15 \(\% \) reduction in error, thus setting a new benchmark in the field.

Knowledge-enhanced online doctor recommendation framework based on knowledge graph and joint learning

A well-performed doctor recommendation system is significant for both patients and Online Medical Consultation Platforms (OMCPs). Though previous studies have proposed many doctor recommendation methods, some are overly personalized for implementation in large-scale OMCPs, while some other machine learning-based approaches perform poorly due to the simplistic information available about patients and doctors on OMCPs. This research proposes an online doctor recommendation framework based on knowledge graph (KG) and joint learning to address these problems. The framework first constructs a comprehensive medical KG, including details about doctors on the platform and a wealth of medical knowledge, to better extract features of doctors and patients. It obtains feature representations of doctors from the medical KG and extracts features from patients’ consultation texts at both sentence and word levels using word embedding and KG embedding. Finally, these features are fed into a deep neural network to calculate the recommendation probability. All processes are learned simultaneously within an overall framework. Extensive experiments conducted on four real datasets illustrate the superior performance of our model and the effectiveness of incorporating KG into doctor recommendation in providing interpretations for the recommendation results.

Learning a Tracking Controller for Rolling μbots.

Micron-scale robots (μbots) have recently shown great promise for emerging medical applications. Accurate control of μbots, while critical to their successful deployment, is challenging. In this work, we consider the problem of tracking a reference trajectory using a μbot in the presence of disturbances and uncertainty. The disturbances primarily come from Brownian motion and other environmental phenomena, while the uncertainty originates from errors in the model parameters. We model the μbot as an uncertain unicycle that is controlled by a global magnetic field. To compensate for disturbances and uncertainties, we develop a nonlinear mismatch controller. We define the model mismatch error as the difference between our model's predicted velocity and the actual velocity of the μbot. We employ a Gaussian Process to learn the model mismatch error as a function of the applied control input. Then we use a least-squares minimization to select a control action that minimizes the difference between the actual velocity of the μbot and a reference velocity. We demonstrate the online performance of our joint learning and control algorithm in simulation, where our approach accurately learns the model mismatch and improves tracking performance. We also validate our approach in an experiment and show that certain error metrics are reduced by up to 40%.

Self-Supervised Self-Organizing Clustering Network: A Novel Unsupervised Representation Learning Method.

Deep learning-based clustering methods usually regard feature extraction and feature clustering as two independent steps. In this way, the features of all images need to be extracted before feature clustering, which consumes a lot of calculation. Inspired by the self-organizing map network, a self-supervised self-organizing clustering network ( [Formula: see text]OCNet) is proposed to jointly learn feature extraction and feature clustering, thus realizing a single-stage clustering method. In order to achieve joint learning, we propose a self-organizing clustering header (SOCH), which takes the weight of the self-organizing layer as the cluster centers, and the output of the self-organizing layer as the similarities between the feature and the cluster centers. In order to optimize our network, we first convert the similarities into probabilities which represents a soft cluster assignment, and then we obtain a target for self-supervised learning by transforming the soft cluster assignment into a hard cluster assignment, and finally we jointly optimize backbone and SOCH. By setting different feature dimensions, a Multilayer SOCHs strategy is further proposed by cascading SOCHs. This strategy achieves clustering features in multiple clustering spaces. [Formula: see text]OCNet is evaluated on widely used image classification benchmarks such as Canadian Institute For Advanced Research (CIFAR)-10, CIFAR-100, Self-Taught Learning (STL)-10, and Tiny ImageNet. Experimental results show that our method significant improvement over other related methods. The visualization of features and images shows that our method can achieve good clustering results.

Joint Learning of Full-Structure Noise in Hierarchical Bayesian Regression Models.

We consider the reconstruction of brain activity from electroencephalography (EEG). This inverse problem can be formulated as a linear regression with independent Gaussian scale mixture priors for both the source and noise components. Crucial factors influencing the accuracy of the source estimation are not only the noise level but also its correlation structure, but existing approaches have not addressed the estimation of noise covariance matrices with full structure. To address this shortcoming, we develop hierarchical Bayesian (type-II maximum likelihood) models for observations with latent variables for source and noise, which are estimated jointly from data. As an extension to classical sparse Bayesian learning (SBL), where across-sensor observations are assumed to be independent and identically distributed, we consider Gaussian noise with full covariance structure. Using the majorization-maximization framework and Riemannian geometry, we derive an efficient algorithm for updating the noise covariance along the manifold of positive definite matrices. We demonstrate that our algorithm has guaranteed and fast convergence and validate it in simulations and with real MEG data. Our results demonstrate that the novel framework significantly improves upon state-of-the-art techniques in the real-world scenario where the noise is indeed non-diagonal and full-structured. Our method has applications in many domains beyond biomagnetic inverse problems.

S$^{3}$M-Net: Joint Learning of Semantic Segmentation and Stereo Matching for Autonomous Driving

S$^{3}$M-Net: Joint Learning of Semantic Segmentation and Stereo Matching for Autonomous Driving

Towards Scalable Multi-View Clustering via Joint Learning of Many Bipartite Graphs

Towards Scalable Multi-View Clustering via Joint Learning of Many Bipartite Graphs

Enabling cross-country learning andexchange to support universal health coverage implementation.

As countries transition from external assistance while pursuing ambitious plans to achieve universal health coverage (UHC), there is increasing need to facilitate knowledge sharing and learning among them. Country-led and country-owned knowledge management is foundational to sustainable, more equitable external assistance for health and is a useful complement to more conventional capacity-building modalities provided under external assistance. In the context of external assistance, few initiatives use country-to-country sharing of practitioner experiences, and link learning to receiving guidance on how to adapt, apply and sustain policy changes. Dominant knowledge exchange processes are didactic, implicitly assuming static technical needs, and that practitioners in low- and middle-income countries require problem-specific, time-bound solutions. In reality, the technical challenges of achieving UHC and the group of policymakers involved continuously evolve. This paper aims to explore factors which are supportive of experience-based knowledge exchange between practitioners from diverse settings, drawing from the experience of the Joint Learning Network (JLN) for UHC-a global network of practitioners and policymakers sharing experiences about common challenges to develop and implement knowledge products supporting reforms for UHC-as an illustration of a peer-to-peer learning approach. This paper considers: (1) an analysis of JLN monitoring and evaluation data between 2020 and 2023 and (2) a qualitative inquiry to explore policymakers' engagement with the JLN using semi-structured interviews (n = 14) with stakeholders from 10 countries. The JLN's experience provides insights to factors that contribute to successful peer-to-peer learning approaches. JLN relies on engaging a network of practitioners with diverse experiences who organically identify and pursue a common learning agenda. Meaningful peer-to-peer learning requires dynamic, structured interactions, and alignment with windows of opportunity for implementation that enable rapid response to emerging and timely issues. Peer-to-peer learning can facilitate in-country knowledge sharing, learning and catalyse action at the institutional and health system levels.

Joint learning of fuzzy embedded clustering and non-negative spectral clustering

Joint learning of fuzzy embedded clustering and non-negative spectral clustering

Joint extraction model of entity relations based on decomposition strategy

Named entity recognition and relation extraction are two important fundamental tasks in natural language processing. The joint entity-relationship extraction model based on parameter sharing can effectively reduce the impact of cascading errors on model performance by performing joint learning of entities and relationships in a single model, but it still cannot essentially get rid of the influence of pipeline models and suffers from entity information redundancy and inability to recognize overlapping entities. To this end, we propose a joint extraction model based on the decomposition strategy of pointer mechanism is proposed. The joint extraction task is divided into two parts. First, identify the head entity, utilizing the positive gain effect of the head entity on tail entity identification.Then, utilize a hierarchical model to improve the accuracy of the tail entity and relationship identification. Meanwhile, we introduce a pointer model to obtain the joint features of entity boundaries and relationship types to achieve boundary-aware classification. The experimental results show that the model achieves better results on both NYT and WebNLG datasets.

Deep unfolding as iterative regularization for imaging inverse problems

Deep unfolding methods have gained significant popularity in the field of inverse problems as they have driven the design of deep neural networks (DNNs) using iterative algorithms. In contrast to general DNNs, unfolding methods offer improved interpretability and performance. However, their theoretical stability or regularity in solving inverse problems remains subject to certain limitations. To address this, we reevaluate unfolded DNNs and observe that their algorithmically-driven cascading structure exhibits a closer resemblance to iterative regularization. Recognizing this, we propose a modified training approach and configure termination criteria for unfolded DNNs, thereby establishing the unfolding method as an iterative regularization technique. Specifically, our method involves the joint learning of a convex penalty function using an input-convex neural network to quantify distance to a real data manifold. Then, we train a DNN unfolded from the proximal gradient descent algorithm, incorporating this learned penalty. Additionally, we introduce a new termination criterion for the unfolded DNN. Under the assumption that the real data manifold intersects the solutions of the inverse problem with a unique real solution, even when measurements contain perturbations, we provide a theoretical proof of the stable convergence of the unfolded DNN to this solution. Furthermore, we demonstrate with an example of magnetic resonance imaging reconstruction that the proposed method outperforms original unfolding methods and traditional regularization methods in terms of reconstruction quality, stability, and convergence speed.

A dual-population multiobjective co-evolutionary matching ensemble learning for product multi-indicator prediction in continuous annealing

Achieving simultaneous and precise prediction of multiple product performance indicators is crucial to ensure stable production in the continuous annealing process. However, the majority of existing quality prediction modeling methods concentrate on developing single-output regression models that are limited to single-indicator prediction scenarios. When dealing with the multi-indicator prediction scenarios, these methods usually require the construction of a separate model for each performance indicator, which leads to more computational and storage resources and more complex maintenance in practical applications. To address this issue, a dual-population multiobjective co-evolutionary matching ensemble learning method called DP-CEMEL is developed. It utilizes dual-population co-optimization to enable the joint learning of the algorithm on input–output relationships and inter-indicator correlations. More specifically, a two-stage learning based individual encoding and decoding method and a performance boosting degree (PBD) based individual evaluation strategy are designed to achieve co-optimization of the dual populations to evolve a set of individuals (i.e., base learners) with good accuracy and low complexity. Then, a PBD based matching ensemble method is applied to get the final multi-indicator prediction model by the optimal matching and ensemble of the evolved individuals. Experimental results on both benchmark data and practical strip production data demonstrate that the proposed DP-CEMEL attains competitive or better performance compared to other state-of-the-art learning methods.

A knowledge-guided graph attention network for emotion-cause pair extraction

Emotion-Cause Pair Extraction (ECPE) is a research objective focused on identifying and extracting all emotion-clause and cause-clause pairs from unannotated emotional text. Traditional methodologies have predominantly employed attention mechanisms or joint learning techniques for feature information interaction. However, these approaches often overlook the aggregation of features under the guidance of external knowledge. To enhance performance in addressing this ECPE challenge, we propose a novel knowledge-guided graph attention network, i.e., GAT-ECPE. This model chiefly relies on an interclause dependency graph as a guiding principle. By employing this knowledge-guided graph attention network, we can proficiently combine semantic and structural information between clauses. In addition, an interpair possibility graph, derived from the outcomes of subtasks, is integrated as an additional guiding principle. As such, we are able to aggregate features between clause pairs, thereby facilitating interaction between multiple tasks. Extensive experiments were conducted to validate our proposed model, and the obtained results demonstrate its superiority when compared to 12 considered baselines. In terms of performance metrics, our model achieves an F1 score F1 of 74.92% and a recall R of 77.52%. These values significantly outperform those achieved by state-of-the-art approaches, indicating the effectiveness and superiority of our GAT-ECPE.

Zero-exemplar deep continual learning for crop disease recognition: a study of total variation attention regularization in vision transformers.

With the increasing integration of AI technology in the food industry, deep learning has demonstrated its immense potential in the domain of plant disease image recognition. However, there remains a gap in research between models capable of continual learning of new diseases and addressing the inherent catastrophic forgetting issue in neural networks. This study aims to comprehensively evaluate various learning strategies based on advanced computer vision models for multi-disease continual learning tasks in food disease recognition. To cater to the benchmark dataset requirements, we collected the PlantDiseaseCL dataset, sourced from the internet, encompassing diverse crop diseases from apples, corn, and more. Utilizing the Vision Transformer (ViT) model, we established a plant disease image recognition classifier, which, in joint learning, outperformed several comparative CNN architectures in accuracy (0.9538), precision (0.9532), recall (0.9528), and F1 score (0.9560). To further harness the potential of ViT in food disease defect recognition, we introduced a mathematical paradigm for crop disease recognition continual learning. For the first time, we proposed a novel ViT-TV architecture in the multi-disease image recognition scenario, incorporating a Total Variation (TV) distance-based loss (TV-Loss) to quantify the disparity between current and previous attention distributions, fostering attention consistency and mitigating the catastrophic forgetting inherent in ViT without prior task samples. In the incremental learning of the PlantDiseaseCL dataset across 3-Steps and 5-Steps, our strategy achieved average accuracies of 0.7077 and 0.5661, respectively, surpassing all compared Zero-Exemplar Approaches like LUCIR, SI, MAS, and even outperforming exemplar-based strategies like EEIL and ICaRL. In conclusion, the ViT-TV approach offers robust support for the long-term intelligent development of the agricultural and food industry, especially showcasing significant applicability in continual learning for crop disease image recognition.