convolutional neural networks

High barriers to adopting pre-trained models and a lack of empirical guidance for strategy selection hinder practical deployment in few-shot image classification and object detection. Method: We systematically compare linear probing versus fine-tuning across ResNet, MobileNet, and EfficientNet, and propose an end-to-end TensorFlow framework integrating multi-scale feature-space visualization (PCA, t-SNE, UMAP) to unify analysis of representation evolution. Contribution/Results: Linear probing significantly outperforms fine-tuning under extreme data scarcity (≤100 samples per class) while accelerating training by 3–5×. The framework enables high-accuracy, rapid deployment (<1 hour for fine-tuning) on standard benchmarks (ImageNet-1K, CIFAR-100), balancing beginner-friendly usability with expert-level extensibility. It bridges the gap between theoretical representation analysis and real-world engineering practice.

To address performance limitations in object detection and semantic segmentation under complex scenarios—including occlusion, small objects, and cross-domain generalization—this paper proposes a novel multimodal detection paradigm synergizing large language models (LLMs). Methodologically, it systematically integrates CNNs, YOLOv5/v8, and DETR architectures into an LLM-augmented inference framework, augmented by scalable data pipelines, model pruning, and quantization, and evaluated via a multi-dimensional metric system based on mAP and mIoU. Key contributions include: (1) bridging the gap between traditional feature engineering and end-to-end deep learning; (2) introducing a dynamic context enhancement mechanism tailored for challenging environments; and (3) achieving state-of-the-art accuracy-efficiency trade-offs on COCO and ADE20K. The fully open-sourced, reproducible framework significantly improves model generalizability and robustness across diverse real-world conditions.

Traditional hand-crafted histogram features—such as Local Binary Patterns (LBP) and edge histograms—are incompatible with end-to-end deep learning due to their non-differentiability. To address this, we propose a differentiable histogram layer, enabling the first neuralization and learnability of such features. Methodologically, we design Neural Local Binary Patterns (NLBP) and Neural Edge Histogram Descriptor (NEHD) modules, integrated as differentiable statistical layers within CNNs to support gradient backpropagation and joint optimization. Our core contribution lies in unifying hand-engineered feature design with deep learning paradigms, allowing local statistical priors to be data-drivenly learned and enhanced. Extensive experiments on multiple image classification benchmarks and real-world datasets demonstrate consistent and significant performance gains, validating that neuralized histogram features substantially improve representation capability.

Neural operators—deep models mapping between function spaces rather than vector spaces—lack open-source, discretization-agnostic implementations with theoretical convergence guarantees. To address this gap, we introduce NeuralOperator, the first modular and extensible Python library for neural operators built on PyTorch. It systematically supports state-of-the-art architectures—including Fourier Neural Operators (FNO) and Multipole Graph Neural Operators (MGNO)—and enables training and inference with functional inputs/outputs under diverse discretizations while rigorously ensuring discretization consistency and convergence. Through a unified API, comprehensive test coverage, and an end-to-end deployment toolchain, NeuralOperator significantly lowers the barrier to adopting neural operators in scientific computing tasks such as partial differential equation solving. The library bridges cutting-edge representational capacity with production-grade engineering robustness, making it both research-ready and deployable in real-world applications.

This work investigates the fundamental differences between spatial token mixers (STMs)—the spatial feature aggregation mechanisms—in Vision Transformers and convolutional networks. To enable a fair, architecture-agnostic comparison, we propose a unified STM modeling paradigm that decouples network-level design from the spatial aggregation module, implementing both convolutional and attention-based STMs on a neutral backbone. Our methodology includes: (1) designing a modular, swappable STM interface; (2) systematically analyzing inductive biases—including receptive field size, translation invariance, and adversarial robustness; and (3) conducting multi-task performance benchmarking. Results show that while modern network-level designs yield substantial gains, intrinsic performance gaps among STMs persist. Crucially, we quantitatively demonstrate for the first time that convolutions exhibit superior translation invariance and local robustness, whereas attention achieves larger effective receptive fields but is more vulnerable to input perturbations.

Existing object detectors often learn task-driven features that rely on shortcut correlations, failing to adequately capture the underlying annotation structure, which limits their generalization, interpretability, and robustness under task shifts or sparse supervision. To address this, this work proposes an annotation-guided feature enhancement framework that explicitly integrates geometric annotation priors into feature learning for the first time. By constructing a dense spatial feature grid and injecting it into the backbone network—where it fuses with the feature pyramid—the method steers region proposal and detection heads toward representations better aligned with annotation structure. Evaluated on wildlife and remote sensing datasets, the approach significantly improves object focus, reduces background sensitivity, and demonstrates superior generalization and data efficiency in weakly supervised and unseen-task settings.

This work proposes a self-supervised feature learning method specifically designed for object detection to address the heavy reliance on large-scale annotated data. By pretraining the feature extractor on unlabeled data and guiding the model to focus on semantically informative object regions, the approach significantly enhances the representational capacity of the detector under limited annotation budgets. Experimental results demonstrate that the proposed method outperforms conventional ImageNet-pretrained models across multiple object detection benchmarks, achieving not only improved detection accuracy but also greater robustness and reliability.

This work proposes a lightweight, customized CNN architecture to address the significant disparities between agricultural and urban scene images—particularly in illumination, resolution, environmental complexity, and class imbalance—and to build an efficient, robust general-purpose visual classification model. Through systematic comparisons with mainstream architectures such as ResNet-18 and VGG-16 across five heterogeneous datasets, the study evaluates the proposed model’s convergence behavior, generalization capability, and performance under both from-scratch training and transfer learning settings across varying data scales. Experimental results demonstrate that the custom CNN achieves accuracy comparable to established models while maintaining a compact footprint. Furthermore, this study provides the first systematic characterization of the practical performance boundaries of transfer learning in small-sample, highly heterogeneous scenarios, offering both theoretical insights and practical guidance for deployment under resource constraints.

This study investigates the generalization capability and overfitting behavior of neural networks on the CIFAR-10 image classification task. By constructing and comparing a fully connected network with a convolutional architecture comprising six convolutional layers and three max-pooling layers, the work implements a complete pipeline encompassing data preprocessing (normalization and one-hot encoding), training (using the Adam optimizer with mini-batches), and validation. After ten training epochs, the model achieves a validation accuracy of 74.77% and clearly exhibits the hallmark overfitting pattern: training loss continues to decrease while validation loss begins to rise. The findings underscore the distinction between representation learning and mere memorization, offering a reproducible benchmark framework that can inform the development of regularization techniques, data augmentation strategies, and educational experimentation.

This work proposes VeloxNet, a lightweight convolutional neural network architecture tailored for embedded image classification, where balancing accuracy and resource constraints is critical. VeloxNet introduces spatial gating units (SGUs) into embedded vision for the first time, replacing the Fire modules in SqueezeNet with gMLP blocks to model global spatial dependencies within a single layer—thereby overcoming the limited receptive fields of conventional local convolutions. By integrating efficient depthwise convolutions with multilayer perceptron structures, VeloxNet achieves notable performance gains: it improves weighted F1 scores by 6.32%, 30.83%, and 2.51% on the AIDER, CDD, and LDD datasets, respectively, while reducing the parameter count by 46.1% compared to SqueezeNet.