With the rapid development of deep learning, a wide range of neural network-based causal discovery frameworks have emerged. Although these methods have achieved significant advancements, they still face several limitations when deployed in real-world industrial processes. Many existing models follow the component-wise modeling design, where an individual model must be built for each variable. This leads to significant computational overhead, especially in high-dimensional industrial systems. Moreover, imposing sparsity constraints on the first-layer weights of neural networks to discover causal relationships limits their ability to capture complex and nonlinear interactions among variables. To address these challenges, we propose a novel lightweight causal discovery framework, termed gradient-based causal discovery (GCD). Different from conventional component-wise models, GCD only employs a single multilayer perceptron for time-series prediction and leverages$\ell _{1}$regularization on the neural network’s input–output gradient to infer causal relationships. Numerical simulations on the Lorenz-96 and CausalTime show that GCD consistently achieves the state-of-the-art performance. Moreover, evaluations across four industrial processes, including Tennessee-Eastman, ultra-processed food, debutanizer, and gas turbine power generation, demonstrate that GCD significantly reduces computational overhead while maintaining high causal discovery accuracy, highlighting its applicability to complex industrial processes.

Hubs are a few points that frequently appear in the k-nearest neighbors (kNN) of many other points in a high-dimensional data set. The hubs' effects, called the hubness phenomenon, degrade the performance of kNN based models in high dimensions. We present SamHub, a simple sampling approach to efficiently identify hubs with theoretical guarantees. Apart from previous works based on approximate kNN indexes, SamHub is generic and applicable to any distance measure with negligible additional memory footprint. Empirically, by sampling only 10% of points, SamHub runs significantly faster and offers higher accuracy than existing hub detection methods on many real-world data sets with dot product, L1, L2, and dynamic time warping distances. Our ablation studies of SamHub on improving kNN-based classification show potential for other high-dimensional data analysis tasks.

Alzheimer's disease (AD) is a progressive neurodegenerative disorder that predominantly affects the elderly population and currently has no cure. Magnetic Resonance Imaging (MRI), as a non-invasive imaging technique, is essential for the early diagnosis of AD. MRI inherently contains both spatial and frequency information, as raw signals are acquired in the frequency domain and reconstructed into spatial images via the Fourier transform. However, most existing AD diagnostic models extract features from a single domain, limiting their capacity to fully capture the complex neuroimaging characteristics of the disease. While some studies have combined spatial and frequency information, they are mostly confined to 2D MRI, leaving the potential of dual-domain analysis in 3D MRI unexplored. To overcome this limitation, we propose Spatial-Frequency Network (SFNet), the first end-to-end deep learning framework that simultaneously leverages spatial and frequency domain information to enhance 3D MRI-based AD diagnosis. SFNet integrates an enhanced dense convolutional network to extract local spatial features and a global frequency module to capture global frequency-domain representations. Additionally, a novel multi-scale attention module is proposed to further refine spatial feature extraction. Experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset demonstrate that SFNet outperforms existing baselines and reduces computational overhead in classifying cognitively normal (CN) and AD, achieving an accuracy of 95.1%.

Intermittent theta burst stimulation (iTBS) is a non-invasive technique frequently employed to induce neural plastic changes and enhance visual attention. Currently, most studies utilized a single iTBS session on healthy subjects to induce short-term neural plastic changes within tens of minutes post-stimulation and investigate its single-session effect on attention performance. Few studies have conducted multiple iTBS sessions on the cerebellum to explore long-term effects on the cerebral cortex and daily effects on visual attention performance. In this study, 18 healthy subjects were involved in a randomized, sham-controlled experiment over one week. All the subjects received daily session of bilateral cerebellar Crus II iTBS or sham stimulation and completed a visual search task. Resting-state electroencephalogram (EEG) was collected 48 hours pre- and post-experiment to assess plastic changes induced by iTBS. The results indicated that the iTBS group exhibited higher accuracy and lower time costs than the sham group after three sessions of iTBS. In addition, iTBS-induced plastic changes persisted up to 48 hours post-experiment, including left-shifted individual alpha frequency, increased intrinsic excitability (the likelihood that a neuron will generate an output in response to a given input), and enhanced PLV functional connectivity (phase synchronization between different brain region). Furthermore, we found that cerebellar iTBS induced a remote effect on the frontal region. Our study revealed the capacity of cerebellar Crus II iTBS to induce plastic changes and enhance attention performance, providing a potential avenue for using iTBS to promote rehabilitation.

In this paper, a low consumption control problem for a class of time-delay systems with sudden non-affine faults is studied. The finite covering lemma and fuzzy logic systems are utilized to eliminate the effect of unknown time delays. The mean value theorem and nussbaum function are used to handle non-affine structures. And then, unknown functions present in the system will be approximated using RBF neural networks. Finally, the event-triggered mechanism is combined with the backstepping framework to achieve low consumption control of the system.

In vector graphics, images are described by mathematical formulas with full image details even after scaling. Most research on generating vector graphics from raster images adopt the approach of splicing graphic fragments, which cannot perfectly retain the original topological structure and details of images. In this paper, we propose TSVec, a novel model for generating high-quality vector graphics by raster images, which takes advantages of the vision transformer and image super-resolution to enhance the granularity of vectorization, especially suitable for dealing with low-resolution raster images. The experimental results show that the vector graphics generated by TSVec outperform the current unsupervised vector generation models.

TSV failures due to manufacturing defects and thermal-induced latent defects result in yield and reliability issues in 3D-ICs. Recent work has shown different temperature mitigation techniques and fault tolerant architectures for 3D-ICs. It is known that TSVs are effective in reducing temperature by providing thermal conductivity. This is the first work that jointly considers temperature mitigation and fault tolerance for TSV based 3D ICs without introducing additional redundant TSVs (also called dummy TSVs). By reusing and carefully placing spare TSVs that are frequently deployed for improving yield and reliability in 3D ICs, temperature is reduced without affecting fault tolerance capability. The proposed technique consists of two steps: first is TSV determination step, which provides optimised allocation of regular and spare TSVs in groups to achieve expected repair capability. The second step is TSV placement, where, for the first time, temperature mitigation is addressed when considering TSVs impact on both vertical and horizontal heat flow. Meanwhile routing difference and total wirelength can be co-optimized. Simulation results show that using the proposed technique, 100% repair capability is achieved across all (five) benchmarks with an average temperature reduction of 33% (best case is 58.2%), while the wirelength may slightly increase depending on assumed TSV fault rate.

Every year many vehicle departure accidents happen due to the driver's carelessness. Lane Departure Warning System (LDWS) is a kind of system which can relieve the stress of the drivers and reduce traffic accidents. But most traffic scenes have greater dynamic range than the digital camera at present. It makes the accuracy of the system would be affected by the complicated lighting. Traditional lane detection methods always use a usual image taken by the camera to detect the lane boundary. In this paper, we will use three images with different exposure to merge a high dynamic range (HDR) image and detect the lane in the HDR image. The experimental results show that the high dynamic range image can improve the accuracy of the lane detection method. However, processing of merging HDR image is very time consuming. It makes HDR image can't be used in real-time LDWS. We proposed an improved method based on exposure fusion to reduce the computational time of the system.