Search Results (2,035)

To address the challenges of limited annotated datasets, strong demand for few-shot adaptation, and insufficient feature representation of traditional backbone networks in Mars rover onboard surface scene classification, this study proposes a classification framework that integrates a channel attention mechanism with meta-transfer learning. The work consisted of two main components. First, a dedicated annotated dataset for core Mars rover onboard surface scenes was constructed based on publicly available Mars Science Laboratory (MSL) Mastcam RGB imagery, providing a standardized experimental benchmark. Second, a classification model was developed based on ResNet50. Transfer learning with ImageNet pre-trained weights was employed to improve feature initialization, and an Squeeze-and-Excitation (SE) attention module was incorporated to enhance channel-wise feature representation. Furthermore, an appropriate activation function was selected under engineering constraints. Finally, a meta-learning strategy based on Model-Agnostic Meta-Learning (MAML) was introduced to improve adaptation capability in few-shot scenarios. Experimental results showed that the SE-ResNet model achieved a validation accuracy of 95.52%, representing an improvement of 17.41% over the baseline model. After integrating meta-transfer learning, the proposed SE-ResNet-MTL (SE-Enhanced ResNet Meta-Transfer Learning) model achieved accuracies of 90.5% and 91.5% in 1-shot and 5-shot tasks, respectively, outperforming traditional fine-tuning methods by 42.8% and 25.4%. These improvements were obtained with only a 5.47% decrease in full-dataset accuracy. Overall, the proposed method effectively balances full-dataset performance and few-shot generalization, providing a practical and efficient solution for Mars rover onboard surface scene recognition and other extraterrestrial visual tasks. Full article

Transcription Factor Binding Site (TFBS) prediction is central to understanding gene regulation and various biological processes. This study introduces HetMoE, a heterogeneous, embedding-gated Mixture-of-Experts for TFBS prediction. A gating network operates on the embeddings produced by a pool of complementary expert backbones (a modified-DeepBIND convolutional network, DeepSEA, and DanQ, with a fine-tuned DNABERT-6 genomic language model as an optional expert), so that models of different architectures are combined and weighted on a per-input basis. Models are trained against GC- and repeat-matched real genomic negatives, a fair protocol that avoids the dinucleotide-shuffle artifact, and evaluated with a balanced-test-set protocol (deterministic inference, $B=1000$ paired bootstrap and Analysis of Variance (ANOVA)) on in-distribution and out-of-distribution (OOD) factors. HetMoE attains the best in-distribution performance (mean Area Under the Curve (AUC) $0.881$) and, on a held-out set stratified by DNA-binding-domain family, surpasses fine-tuned DNABERT-6 on the motif-bearing OOD mean across three random seeds ($0.821\pm 0.005$ vs. $0.799\pm 0.008$, a gain present in every seed), most strongly on the sequence-specific and within-family factors. The advantage comes from the gating mechanism rather than from ensembling: input-dependent gating exceeds a static average of the same experts by $0.073$ AUC and the best single expert by $0.088$, and the configuration selected on in-distribution data is a pretraining-free pool of convolutional experts. We further show that the common dinucleotide-shuffle negative protocol inflates the apparent margin (to a mean of $0.864$), which shows the importance of fair, genomically matched negatives. We also introduce an attribution method (ShiftSmooth) that improves interpretability by averaging the gradient over small shifts of the input sequence, giving more reliable attribution for motif discovery and localization than the Vanilla Gradient. Together these provide an efficient and interpretable approach to TFBS prediction that can support further study of genome regulation. Full article

Low-altitude radar target recognition is important for intelligent airspace monitoring, unmanned aerial vehicle (UAV) supervision, airport bird-strike prevention, and low-altitude remote sensing. Reliable recognition remains difficult because birds, balloons, and UAVs often produce weak radar responses, share similar trajectory-level signatures, and are difficult to annotate at scale. To address these challenges, this paper proposes a physics-informed semantic prompt learning framework for few-shot low-altitude radar target recognition. The framework converts radar point-track and track measurements into structured textual prompts that combine statistical descriptors, radar-domain physical knowledge, and task-specific instructions. A partially fine-tuned Generative Pre-trained Transformer 2 (GPT-2) encoder is then used to extract semantic representations that preserve motion and scattering-related information. An adaptive feature aggregation module further weights informative hidden states across temporal positions and semantic levels, and a relation-based meta-learning network models query-support similarity for few-shot classification. Experiments on a real low-altitude radar dataset with four target categories, namely birds, balloons, small rotary-wing UAVs, and light rotary-wing UAVs, show that the proposed method consistently outperforms conventional machine learning, deep learning, and representative few-shot baselines. Under the 20-shot setting, it achieves mean 90.85% precision, 90.47% recall, and 90.63% F1-score. The results indicate that embedding radar physical semantics into language-model-based representation learning can improve sample efficiency and recognition robustness for low-altitude radar remote sensing. Full article

The ubiquitin–proteasome system (UPS) has traditionally been described as a tightly regulated degradative network driven mainly by the specificity of its ubiquitin-conjugating enzymatic components. The 26S proteasome is the catalytic arm of the system that acts downstream to the conjugation machinery. For a long time, it has been considered to be a constitutive multi-subunit proteolytic complex that recognizes in a non-discriminatory manner ubiquitin-marked target substrates with less than a handful of exceptions. However, emerging evidence reveals that the 26S proteasome function is also dynamically regulated by multiple factors, such as subunit composition and synthesis, post-translational modifications, and spatial localization, all of which are tightly regulated by the metabolic and stress states of the cell. Importantly, dysregulation of these newly emerging regulatory mechanisms has pathogenic sequelae. These mechanisms fine-tune proteasome activity and expand its role as an active regulator of protein homeostasis rather than being a passive degradation machinery. Given the rapid expansion of these findings and their impact on our understanding of proteasome biology, an integrated overview of these regulatory mechanisms is timely. Full article

Balancing detection accuracy with model lightweightness remains a key challenge in remote sensing object detection. Although convolutional neural networks have improved performance, they typically require large-scale datasets, making few-shot detection of novel classes difficult. To tackle this, we propose LFODet, a lightweight few-shot object detection network based on meta-learning. It uses two parallel branches to rapidly adapt to novel classes with limited samples while maintaining performance on base classes. For efficient feature representation, we integrate Semantic Ghost Channel Attention (GCA) and Fine-Grained Ghost Spatial Attention (GSA) to enhance semantic discriminability and spatial detail preservation. Moreover, we leverage Ghost convolutions to reduce computational complexity. The model is trained in three stages: base-class pre-training, meta-learner optimization, and few-shot fine-tuning. Experiments on DIOR and NWPU VHR-10 demonstrate that LFODet achieves stable and balanced performance across various few-shot learning scenarios. As validated on these benchmark datasets, this work provides a practical solution for resource-constrained remote sensing applications requiring rapid adaptation to new targets. Full article

Compared with traditional artificial neural networks (ANNs), spiking neural networks (SNNs) have lower computational complexity, lower energy consumption, and faster inference speed, making them more promising for practical deployment on edge devices. However, in SNNs, since the output spikes of neurons are discrete, the model may face information loss, especially when the membrane potential is quantized into binary spikes, where quantization errors can lead to model precision loss and information loss. To address these challenges, this study proposes a spiking neural network based on hierarchical residual quantization and temporal error compensation (HRQ-TEC-SNN) and used for remote sensing object detection tasks. Through hierarchical residual quantization and temporal error compensation design, higher resolution quantization of membrane potential can be performed and the quantization threshold of membrane potential can be dynamically adjusted to compensate for errors introduced during the quantization process, allowing for fine-tuning at each time step and reducing information loss caused by coarse quantization. In terms of network structure, by introducing depthwise separable convolution modules, channel attention and spatial attention mechanisms, and integrating fast spatial pyramid pooling based on pulse neural networks, the model’s detection accuracy has been further improved while reducing the number of model parameters and computational costs. Experimental results show that the HRQ-TEC-SNN achieves significant advantages in both accuracy and energy consumption on the DOTAv1.0, DOTAv1.5 and DIOR datasets. Full article

Lightweight, privacy-aware and adversarial robust intrusion detection is required for the proliferation of Internet of Things (IoT) devices. In the Industrial Internet of Things (IIoT), centralized detectors can be compromised by adversarial perturbations via gradient-based attacks, making them susceptible to raw traffic. We suggest Federated Learning-Adaptive Gated Recurrent Unit (FL-AdGRU), a Federated approach that combines a lightweight Gated Recurrent Unit (GRU) classifier with alternating adversarial fine-tuning on each client using FGSM and PGD, without any communication overhead. A two-stage resampling scheme (UCAS-SMOTE) reduces the class-imbalance ratio from 4081:1 to ≈4:1, followed by 61 features being reduced to 40 by a mutual-information selector (MI-SelectK). Under this scenario, FL-AdGRU achieves 99.9% accuracy and 0.999 weighted F1 (+6.5 p.p. over the federated DNN baseline), with no loss of accuracy when facing clean attacks, and boosts Fast Gradient Sign Method FGSM/Projected Gradient Descent (PGD) robustness by +19.3/+19.0 p.p. at the same level of $ϵ$ = 0.1, thus effectively balancing the accuracy–robustness trade-off. It is robust (97.8%/84.2% on UNSW-NB15) and generalizes well to UNSW-NB15, while decaying slowly in skeptical scenarios (≈99.9% weighted F1 for moderate skew, 93.9%/86.7% for severe). Assuring data-locality privacy through exchange of only model weights; defenses against inference attack are left for future work. FL-AdGRU, with a total communication of 43.8 MB (≈50× less than centralized training), is deployable on bandwidth-constrained IIoT networks. Full article

State-of-the-art generative models exhibit powerful image-generation capabilities, raising ethical and legal challenges for service providers. Consequently, Content Removal Techniques (CRTs) have emerged to control outputs without requiring full retraining. However, the problem of unlearning in Generative Adversarial Networks (GANs) remains largely unexplored. We propose Text-to-Unlearn, a novel framework that selectively unlearns concepts from pre-trained GANs using only text prompts, enabling feature and identity unlearning, as well as fine-grained tasks such as expression and multi-attribute removal in models trained on human faces. Our approach leverages natural language descriptions to guide unlearning without additional datasets or supervised finetuning, offering a scalable solution. To evaluate the effectiveness of our method, we introduce an automated unlearning assessment method using state-of-the-art image–text alignment metrics and propose a new metric: degree of unlearning. Additionally, we assess robustness by introducing adversarial attacks to subvert unlearning. Our results demonstrate that Text-to-Unlearn achieves robust unlearning, resisting adversarial attempts to recover erased concepts while preserving model utility. To our knowledge, this is the first cross-modal unlearning framework for GANs, advancing the management of generative model behavior. Full article

Background/Objectives: Skin cancer is one of the most common diseases worldwide, with a high mortality rate. Due to its ability to metastasize, the disease can progress to more serious stages over time. This article proposes a hybrid model based on feature engineering that will play a critical role in the early diagnosis of the disease. Methods: The developed model in this paper utilizes the well-known Vision Transformer (ViT) and Convolutional Neural Network (CNN) models for feature extraction from images in the dataset, while the FT-Transformer, Excel Former, SAINT, GRANDE, PTaRL, and TabTransformer architectures are used for feature extraction from clinical data. Furthermore, this study was developed using a very large pool of classifiers, including 13 classifiers. Fine-tuning was applied to improve the performance of the developed model. Channel attention mechanisms were incorporated into the study to ensure that the proposed model focuses on the diseased area. The PAD-UFES-20 dataset was used during the experiments. Class weighting was applied to the proposed model to prevent class-based imbalance in the PAD-UFES-20 dataset. Results: Six distinct CNN and four distinct ViT models were compared to the developed model. The developed model achieved a highly competitive Area Under the Curve (AUC) rate of 96.41%. The study was conducted using a dataset containing both clinical and imaging data. Conclusions: The proposed model is thought to help dermatologists diagnose skin cancer. Full article

Accurate change detection in high-resolution remote sensing imagery is essential for urban planning, land-use monitoring, and disaster response. This study introduces CTA-Net, a Cross-Temporal Attention Network for binary change detection in bi-temporal optical imagery, designed to improve robustness against pseudo-changes caused by illumination variation, seasonal effects, and sensor noise. The proposed method employs a shared Siamese encoder with multi-scale Cross-Temporal Attention modules that derive spatial and channel attention from L2 feature differences, along with a lightweight confidence estimation head for per-pixel uncertainty modelling. A hybrid loss function combining confidence-weighted binary cross-entropy and focal loss is used to address class imbalance. Experiments on the LEVIR-CD dataset demonstrate that CTA-Net achieves an overall accuracy of 98.99%, an F1-score of 87.68%, an Intersection over Union of 78.06%, a Cohen’s kappa of 0.8715, and a Matthews Correlation Coefficient of 0.8721, with stable convergence and minimal overfitting. Qualitative and calibration analyses further indicate that the model produces interpretable attention maps and reliable probabilistic outputs. To evaluate cross-domain generalization, we conduct a transfer learning case study on multispectral Sentinel-2 agricultural imagery. The model is adapted to 11-channel input and fine-tuned on automatically generated change masks derived from NDVI-delta thresholding. Under this supervision protocol, CTA-Net achieves an F1-score of 95.18% and an IoU of 90.81% on a held-out test region, with balanced precision and recall. While these results demonstrate effective adaptation across sensor modality, spatial resolution, and semantic domain, the evaluation reflects agreement with the mask generation procedure rather than independently annotated ground truth. While CTA-Net shows strong performance and reasonable interpretability, its cross-domain evaluation is limited by the use of automatically generated labels. As a result, the reported transferability should be interpreted cautiously until validated on human-annotated datasets. Full article

Hellenic Open University has developed Onlabs, a virtual biology laboratory for its students to be trained before they use its on-site lab. The evaluation of the user’s performance in the virtual lab with respect to a particular experimental procedure is done with a scoring algorithm specifically designed for this purpose. For the calculation of the user’s overall progress score, an Artificial Neural Network (ANN) is used. The ANN, trained with data from random plays evaluated by biology experts, achieves significant convergence. Yet, when the trained ANN is used for the real-time evaluation of the user’s performance, it produces unrealistic scores, that is, incompatible with human experience, such as unscaled score values as well as a high increase in score with the execution of secondary actions. To overcome this problem, the ANN’s weights are fine-tuned with the use of a Genetic Algorithm (GA) and two algorithms of Swarm Intelligence (SI), Whale Optimization Algorithm (WOA) and Firefly Algorithm (FA). Among those, GA achieves successful optimization of the ANN’s weights, resulting in a more realistic score mechanism. Full article

Background: Strategic resource planning is crucial for optimizing supply chain management and ensuring efficient operations. This study aims to enhance strategic planning in gold mines by leveraging advanced gold price forecasting models. By predicting future gold prices accurately, mining companies can better plan their extraction, processing, and distribution activities, thereby improving overall supply chain efficiency. Methods: Various advanced forecasting models are implemented, including backpropagation neural networks (BPNNs), convolutional neural network (CNN), long short-term memory (LSTM), bi-directional LSTM (Bi-LSTM), gated recurrent unit (GRU), and bi-directional GRU (Bi-GRU). The feature selection process is facilitated by recursive feature elimination (RFE), and Optuna is used to fine-tune neural network models. Evaluation is based on root mean squared error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). Results: BPNN performed the best in terms of lowest RMSE (0.5928), MAE (0.4091), and MAPE (0.34%), whilst Bi-GRU was the poorest performer, as it achieved RMSE of 9.41, MAE of 8.1916, and MAPE of 6.94%. In addition, Optuna further improved each model’s accuracy, except CNN, where the performance slightly decreased. Conclusions: Advanced forecasting neural systems underperformed the standard backpropagation neural networks. In this regard, BPNN proved to be highly effective in forecasting gold price, providing critical managerial implications for navigating the dynamic and volatile gold market for gold mining companies and investors. Full article

Urban Green Spaces (UGSs) monitoring usually focuses on the extraction of vegetation in the physical layer, while neglecting their functional attributes. This renders the monitoring results unable to objectively reflect the rationality of UGS planning. To address these issues, this study proposes a multi-level classification method integrating multi-source remote sensing and geospatial big data to bridge the semantic gap between the physical layer and the functional layer. In this method, a strategy of prior knowledge injection and semantic reconstruction was developed through the fine-tuning of a BERT model with cross-mapping rules. This strategy aims to classify the urban area into 24 functional categories, generating the social-functional basemap in a functional layer, based on Point of Interest (POI), OpenStreetMap (OSM), and Global Urban Boundary (GUB). Meanwhile, a novel deep learning architecture, namely the Multi-Shape and Spectral Aware Network (MSSANet), was designed for precise vegetation classification of UGSs in the physical layer. Finally, a “function-first, vegetation-second” coupling paradigm containing three functional attribute layers, referring to the Code for Classification of UGS in China (CJJ/T 85-2017), was established. This paradigm integrates the social-functional basemap with physical vegetation patches to build a multi-level UGS classification framework, i.e., the 5 major UGS categories, 11 intermediate UGS categories, and 24 fine-grained UGS sub-categories. Experiments conducted in Jinan and Qingdao, China, demonstrate the efficacy of the proposed method for refined multi-level UGS mapping. Full article

Background: Metastatic lymph node (MLN) detection remains a major clinical challenge in head and neck cancer, as nodal involvement is strongly associated with poor prognosis and directly affects treatment planning. Previous approaches typically rely on cropped lymph node (LN) regions or tumor contours for MLN identification, requiring substantial expert annotation during preprocessing and relying solely on imaging information. As a result, small or low-contrast metastatic nodes may be missed, while benign lymph nodes may be incorrectly identified as metastatic due to overlapping imaging characteristics. To address these limitations, we propose a multimodal learning framework that integrates anatomical and metabolic features from head and neck PET/CT images with biochemical features derived from FLIm for patient-level MLN prediction, without requiring manual lymph node cropping or tumor contouring during inference. Methods: To enable robust imaging representation learning, a region-aware PET/CT network based on a merging-diverging architecture was first pretrained on the HECKTOR 2022 dataset and then fine-tuned on the institutional cohort. In parallel, FLIm point-wise measurements with clinical variables were encoded using a multilayer perceptron (MLP) and aggregated into subject-level representations. To effectively combine these modalities, two multimodal fusion strategies were evaluated at the decoder stage, including cube-based fusion and squeeze-and-excitation (SE)-based fusion. The proposed strategies were evaluated on a cohort of 53 patients. Results: Compared with the single-modality baselines, both multimodal fusion strategies achieved better patient-level MLN prediction. The PET/CT-only segmentation-driven model and FLIm-only model reached balanced accuracies of 0.815 and 0.665, with AUCs of 0.828 and 0.614, respectively. Cube-based fusion improved balanced accuracy and AUC to 0.827 and 0.850, respectively, while channel-wise SE-based fusion achieved the best overall performance, with a balanced accuracy of 0.839 and an AUC of 0.872. Conclusions: These results suggest that multimodal integration may improve patient-level MLN prediction compared with single-modality approaches. Given the limited sample size, these findings should be interpreted as hypothesis-generating and require validation in larger, independent patient cohorts. Full article

South Africa faces significant challenges in monitoring air pollution from different provinces due to the sparse nature of the sensor network and heterogeneous pollutant sources. Notably, some provinces continue to record a limited amount of data on air pollution, thus making monitoring in those locations problematic. Fortunately, the capabilities of deep learning models to facilitate effective monitoring in data-scarce locations have been highlighted by researchers; however, these models within the context of transfer learning still lack transparency and uncertainty quantification. Using air pollutants and meteorological factors, this study proposes a transfer learning model for particulate matter (PM_2.5) prediction in a data-scarce region. This transfer learning (TL) model leverages an adaptive Bi-directional Gated Recurrent Unit (adaBiGRU) with explainable artificial intelligence (xAI) and uncertainty quantification (UQ) to provide a novel uncertainty-aware adaptation transfer learning (UATL_adaBiGRU) model for a data-scarce location. Variant models based on the adaBiGRU technique, such as the temporal convolution network adaBiGRU (TCN-adaBiGRU) and domain-adversarial neural network adaBiGRU (DANNadaBiGRU), are presented as comparative models. The performance evaluation metrics are root mean squared, R² score and mean squared error. The R² score of pre-trained models in source domain is adaBiGRU (0.888), DANN_adaBiGRU (0.7788) and TCN_adaBiGRU (0.876). Furthermore, other comparative TL models include GRU (0.898), MLP (0.802) and adaptive LSTM (0.886). Afterwards, the pre-trained baseline model (adaBiGRU) was fine-tuned in the target domain dataset and the unpromising result contributed to the proposition of the UATL_adaBiGRU model for a data-scarce location, with R² score of 0.9618. Uncertainty assessment metrics results were also presented for the proposed model. Ablation assessment demonstrates that each component of the UATL_adaBiGRU contributes to enhancing the predictive performance. Again, the Diebold–Mariano (DM) test statistic demonstrates a statistically significant difference between baseline model and UATL_adaBiGRU model. Finally, the local interpretable model-agnostic explanation highlights multi-scaled features as contributing towards the prediction of PM_2.5 in the target domain. In view of this result, model fine-tuning is strongly recommended to enhance the robustness of the proposed uncertainty-aware adaption model in data-limited regions in South Africa. Full article