Labeled data
View article: Estimation and Inference for Density-convoluted Support Vector Machine with Streaming Data
Estimation and Inference for Density-convoluted Support Vector Machine with Streaming Data Open
View article: Federated weakly-supervised representation learning for privacy-preserving human activity recognition using wearable sensors
Federated weakly-supervised representation learning for privacy-preserving human activity recognition using wearable sensors Open
Human activity recognition (HAR) plays a crucial role in wearable sensor-based applications, particularly in privacy-sensitive domains such as healthcare, fitness, and smart homes. Federated learning (FL), when combined with weakly-supervi…
View article: Federated weakly-supervised representation learning for privacy-preserving human activity recognition using wearable sensors
Federated weakly-supervised representation learning for privacy-preserving human activity recognition using wearable sensors Open
Human activity recognition (HAR) plays a crucial role in wearable sensor-based applications, particularly in privacy-sensitive domains such as healthcare, fitness, and smart homes. Federated learning (FL), when combined with weakly-supervi…
View article: A Survey of Classification Algorithms in Supervised Machine Learning
A Survey of Classification Algorithms in Supervised Machine Learning Open
Machine learning is crucial in enhancing predictive and diagnostic capabilities across multiple sectors. Professionals can use it to identify potential conditions and assess the risks associated with different interventio…
View article: A Survey of Classification Algorithms in Supervised Machine Learning
A Survey of Classification Algorithms in Supervised Machine Learning Open
Machine learning is crucial in enhancing predictive and diagnostic capabilities across multiple sectors. Professionals can use it to identify potential conditions and assess the risks associated with different intervention strategies. Mach…
View article: A Multi-Task Deep Learning Framework with Clinical Knowledge–Guided Regularization for Concurrent Prediction of Diabetes and Hypertension
A Multi-Task Deep Learning Framework with Clinical Knowledge–Guided Regularization for Concurrent Prediction of Diabetes and Hypertension Open
The concurrent prevalence of diabetes and hypertension poses a major challenge in predictive healthcare, as both conditions share intertwined physiological and metabolic pathways. However, most existing predictive models treat them in isol…
View article: Data-Efficient Deep Learning Framework for Urolithiasis Detection Using Transfer and Self-Supervised Learning
Data-Efficient Deep Learning Framework for Urolithiasis Detection Using Transfer and Self-Supervised Learning Open
Purpose: Recent studies on urolithiasis detection using deep learning have demonstrated promising accuracy; however, most rely on large-scale labeled imaging datasets. In clinical practice, only limited and partially labeled computed tomog…
View article: Understanding generative AI output with embedding models
Understanding generative AI output with embedding models Open
Constructing high-quality features is critical to any quantitative data analysis. While feature engineering was historically addressed by carefully handcrafting data representations on the basis of domain expertise, deep neural networks (D…
View article: Active Learning for GCN-based Action Recognition
Active Learning for GCN-based Action Recognition Open
Despite the notable success of graph convolutional networks (GCNs) in skeleton-based action recognition, their performance often depends on large volumes of labeled data, which are frequently scarce in practical settings. To address this l…
View article: Timestamp Supervision for Surgical Phase Recognition Using Semi-Supervised Deep Learning
Timestamp Supervision for Surgical Phase Recognition Using Semi-Supervised Deep Learning Open
Surgical Phase Recognition (SPR) enables real-time, context-aware assistance during surgery, but its use remains limited by the cost and effort of dense video annotation. This study presents a Semi-Supervised Deep Learning framework for SP…
View article: FedAPA: Federated Learning with Adaptive Prototype Aggregation Toward Heterogeneous Wi-Fi CSI-based Crowd Counting
FedAPA: Federated Learning with Adaptive Prototype Aggregation Toward Heterogeneous Wi-Fi CSI-based Crowd Counting Open
Wi-Fi channel state information (CSI)-based sensing provides a non-invasive, device-free approach for tasks such as human activity recognition and crowd counting, but large-scale deployment is hindered by the need for extensive site-specif…
View article: BanglaMM-Disaster: A Multimodal Transformer-Based Deep Learning Framework for Multiclass Disaster Classification in Bangla
BanglaMM-Disaster: A Multimodal Transformer-Based Deep Learning Framework for Multiclass Disaster Classification in Bangla Open
Natural disasters remain a major challenge for Bangladesh, so real-time monitoring and quick response systems are essential. In this study, we present BanglaMM-Disaster, an end-to-end deep learning-based multimodal framework for disaster c…
View article: Few-shot cross-domain fault diagnosis via adversarial meta-learning
Few-shot cross-domain fault diagnosis via adversarial meta-learning Open
This paper introduces a fault diagnosis approach, MLAML, which reconstructs data reconstruction, meta-learning, and adversarial learning for cross-domain fault diagnosis in small-sample scenarios. To enhance signal quality, an improved spa…
View article: The miniJPAS and J-NEP surveys: Machine learning for star-galaxy separation
The miniJPAS and J-NEP surveys: Machine learning for star-galaxy separation Open
We present a supervised machine learning classification of sources from the Javalambre Physics of the Accelerating Universe Astrophysical Survey (J-PAS) Pathfinder datasets: miniJPAS and J-NEP. Leveraging crossmatches with spectroscopic an…
View article: DiCaP: Distribution-Calibrated Pseudo-labeling for Semi-Supervised Multi-Label Learning
DiCaP: Distribution-Calibrated Pseudo-labeling for Semi-Supervised Multi-Label Learning Open
Semi-supervised multi-label learning (SSMLL) aims to address the challenge of limited labeled data in multi-label learning (MLL) by leveraging unlabeled data to improve the model's performance. While pseudo-labeling has become a dominant s…
View article: Ranking-Enhanced Anomaly Detection Using Active Learning-Assisted Attention Adversarial Dual AutoEncoders
Ranking-Enhanced Anomaly Detection Using Active Learning-Assisted Attention Adversarial Dual AutoEncoders Open
Advanced Persistent Threats (APTs) pose a significant challenge in cybersecurity due to their stealthy and long-term nature. Modern supervised learning methods require extensive labeled data, which is often scarce in real-world cybersecuri…
View article: Bridging the Language Gap: Synthetic Voice Diversity via Latent Mixup for Equitable Speech Recognition
Bridging the Language Gap: Synthetic Voice Diversity via Latent Mixup for Equitable Speech Recognition Open
Modern machine learning models for audio tasks often exhibit superior performance on English and other well-resourced languages, primarily due to the abundance of available training data. This disparity leads to an unfair performance gap f…
View article: Ranking-Enhanced Anomaly Detection Using Active Learning-Assisted Attention Adversarial Dual AutoEncoders
Ranking-Enhanced Anomaly Detection Using Active Learning-Assisted Attention Adversarial Dual AutoEncoders Open
Advanced Persistent Threats (APTs) pose a significant challenge in cybersecurity due to their stealthy and long-term nature. Modern supervised learning methods require extensive labeled data, which is often scarce in real-world cybersecuri…
View article: Multi-label Classification Algorithm for Financial Texts Based on Deep Learning
Multi-label Classification Algorithm for Financial Texts Based on Deep Learning Open
The multi-label classification algorithm for financial texts can implement information retrieval from vast financial data according to user needs. To further enhance the ability to identify financial text labels, this paper proposes a deep…
View article: A novel approach to identify deepfake text using social media data
A novel approach to identify deepfake text using social media data Open
The proliferation of manipulated content, such as counterfeit films, text, audio, and photographs, has surged in recent years due to advanced digital manipulation tools and techniques. Social media platforms are also plagued by false infor…
View article: DiCaP: Distribution-Calibrated Pseudo-labeling for Semi-Supervised Multi-Label Learning
DiCaP: Distribution-Calibrated Pseudo-labeling for Semi-Supervised Multi-Label Learning Open
Semi-supervised multi-label learning (SSMLL) aims to address the challenge of limited labeled data in multi-label learning (MLL) by leveraging unlabeled data to improve the model's performance. While pseudo-labeling has become a dominant s…
View article: Rethinking Semi-Supervised Node Classification with Self-Supervised Graph Clustering
Rethinking Semi-Supervised Node Classification with Self-Supervised Graph Clustering Open
The emergence of graph neural networks (GNNs) has offered a powerful tool for semi-supervised node classification tasks. Subsequent studies have achieved further improvements through refining the message passing schemes in GNN models or ex…
View article: Deep Learning-Based SAR Target Recognition: A Dual-Perspective Survey of Closed Set and Open Set
Deep Learning-Based SAR Target Recognition: A Dual-Perspective Survey of Closed Set and Open Set Open
Owing to the all-weather, day-and-night imaging capability of Synthetic Aperture Radar (SAR), SAR automatic target recognition (ATR) has long been a central focus in academia and industry. Since 2013, deep learning has become the dominant …
View article: Rethinking Semi-Supervised Node Classification with Self-Supervised Graph Clustering
Rethinking Semi-Supervised Node Classification with Self-Supervised Graph Clustering Open
The emergence of graph neural networks (GNNs) has offered a powerful tool for semi-supervised node classification tasks. Subsequent studies have achieved further improvements through refining the message passing schemes in GNN models or ex…
View article: Cross-Contrastive Clustering for Multimodal Attributed Graphs with Dual Graph Filtering
Cross-Contrastive Clustering for Multimodal Attributed Graphs with Dual Graph Filtering Open
Multimodal Attributed Graphs (MMAGs) are an expressive data model for representing the complex interconnections among entities that associate attributes from multiple data modalities (text, images, etc.). Clustering over such data finds nu…
View article: Cross-Contrastive Clustering for Multimodal Attributed Graphs with Dual Graph Filtering
Cross-Contrastive Clustering for Multimodal Attributed Graphs with Dual Graph Filtering Open
Multimodal Attributed Graphs (MMAGs) are an expressive data model for representing the complex interconnections among entities that associate attributes from multiple data modalities (text, images, etc.). Clustering over such data finds nu…
View article: Ranking-enhanced anomaly detection using Active Learning-assisted Attention Adversarial Dual AutoEncoder
Ranking-enhanced anomaly detection using Active Learning-assisted Attention Adversarial Dual AutoEncoder Open
Advanced Persistent Threats (APTs) pose a significant challenge in cybersecurity due to their stealthy and long-term nature. Modern supervised learning methods require extensive labeled data, which is often scarce in real-world cybersecuri…
View article: How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM Pretraining
How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM Pretraining Open
Due to the scarcity of high-quality data, large language models (LLMs) are often trained on mixtures of data with varying quality levels, even after sophisticated data curation. A natural approach to better leverage high-quality data is cu…
View article: Sampling Control for Imbalanced Calibration in Semi-Supervised Learning
Sampling Control for Imbalanced Calibration in Semi-Supervised Learning Open
Class imbalance remains a critical challenge in semi-supervised learning (SSL), especially when distributional mismatches between labeled and unlabeled data lead to biased classification. Although existing methods address this issue by adj…
View article: From Pixels to Posts: Retrieval-Augmented Fashion Captioning and Hashtag Generation
From Pixels to Posts: Retrieval-Augmented Fashion Captioning and Hashtag Generation Open
This paper introduces the retrieval-augmented framework for automatic fashion caption and hashtag generation, combining multi-garment detection, attribute reasoning, and Large Language Model (LLM) prompting. The system aims to produce visu…