Non-Intrusive Load Monitoring (NILM) is widely employed to disaggregate a building’s total electrical load and estimate the energy consumption of individual devices. Recently, image-based NILM has garnered interest for its ability to capture temporal patterns in time series data. However, existing methods often convert time series data into a single image, resulting in information loss and inferior disaggregation performance. To address this issue, we propose a novel tensor-based image encoding approach for NILM. Our method leverages a proposed diagonal projection method, which enables nearly 100% recovery of raw time series values through inverse normalization. It only performs linear scaling without altering inherent data characteristics. We also integrate three additional image conversion techniques-Recurrence Plot, Gramian Angular Field and Markov Transition Field-to construct a four-dimensional image set that preserves data integrity. This set is then encoded into a multi-dimensional tensor, providing rich geometric features for model training. Additionally, we introduce an involution model to expand the convolutional receptive field and reduce parameter redundancy. Experimental results on the REDD and AMPds datasets demonstrate that our approach outperforms existing NILM techniques, highlighting its significant potential for building energy disaggregation.

A common characteristic in integer linear programs (ILPs) is symmetry, allowing variables to be permuted without altering the underlying problem structure. Recently, GNNs have emerged as a promising approach for solving ILPs. However, a significant challenge arises when applying GNNs to ILPs with symmetry: classic GNN architectures struggle to differentiate between symmetric variables, which limits their predictive accuracy. In this work, we investigate the properties of permutation equivalence and invariance in GNNs, particularly in relation to the inherent symmetry of ILP formulations. We reveal that the interaction between these two factors contributes to the difficulty of distinguishing between symmetric variables. To address this challenge, we explore the potential of feature augmentation and propose several guiding principles for constructing augmented features. Building on these principles, we develop an orbit-based augmentation scheme that first groups symmetric variables and then samples augmented features for each group from a discrete uniform distribution. Empirical results demonstrate that our proposed approach significantly enhances both training efficiency and predictive performance.

Pathology image matching is crucial for assisting pathologists in the comprehensive diagnosis of cancerous areas. However, variations in image rotation and staining caused by inherent slide imaging techniques increase the burden on pathologists, complicating the examination of cancer across different pathology slides. To address this challenge, we introduce multi-scale ring rotation-invariant matching (MRRM), which improves image matching efficiency using ring topology, assisting pathologists in robustly aligning biomarker information across various pathology images. Specifically, by employing multi-scale rings as convolution kernels, we accurately locate keypoints from the differencing of the ring pyramid, which not only enhances the likelihood of successful pathology image matching but also supports our feature descriptor in achieving advantageous performance in rotation-invariance. Experiments show that with manually annotated golden landmarks as the standard in 81 cases, exhibiting significantly superior matching accuracy (130.93 $\,\mu \mathrm{m}$) and a success rate of 93.83% compared to other methods, particularly in cases with rotated pathology images. This meets the routine diagnostic requirements of pathologists for cancer diagnosis.

Matching whole slide histopathology images to provide comprehensive information on homologous tissues is beneficial for cancer diagnosis. However, the challenge arises with the Giga-pixel whole slide images (WSIs) when aiming for high-accuracy matching. Learning-based methods are difficult to generalize well with large-size WSIs, necessitating the integration of traditional matching methods to enhance accuracy as the size increases. In this paper, we propose a multi-size guiding matching method applicable high-accuracy requirements. Specifically, we design learning multiscale texture to train deep descriptors, called TDescNet, that trains 64 × 64 × 256 and 256 × 256 × 128 size convolution layer as C64 and C256 descriptors to overcome staining variation and low visibility challenges. Furthermore, we develop the 3D-ring descriptor using sparse keypoints to support the description of large-size WSIs. Finally, we employ C64, C256, and 3D-ring descriptors to progressively guide refined local matching, utilizing geometric consistency to identify correct matching results. Experiments show that when matching WSIs of size 4096 × 4096 pixels, our average matching error is 123.48 μm and the success rate is 93.02 % in 43 cases. Notably, our method achieves an average improvement of 65.52 μm in matching accuracy compared to recent state-of-the-art methods, with enhancements ranging from 36.27 μm to 131.66 μm. Therefore, we achieve high-fidelity whole-slice image matching, and overcome staining variation and low visibility challenges, enabling assistance in comprehensive cancer diagnosis through matched WSIs.

Once challenged by the SARS-CoV-2 virus, the human host immune system triggers a dynamic process against infection. We constructed a mathematical model to describe host innate and adaptive immune response to viral challenge. Based on the dynamic properties of viral load and immune response, we classified the resulting dynamics into four modes, reflecting increasing severity of COVID-19 disease. We found the numerical product of immune system's ability to clear the virus and to kill the infected cells, namely immune efficacy, to be predictive of disease severity. We also investigated vaccine-induced protection against SARS-CoV-2 infection. Results suggested that immune efficacy based on memory T cells and neutralizing antibody titers could be used to predict population vaccine protection rates. Finally, we analyzed infection dynamics of SARS-CoV-2 variants within the construct of our mathematical model. Overall, our results provide a systematic framework for understanding the dynamics of host response upon challenge by SARS-CoV-2 infection, and this framework can be used to predict vaccine protection and perform clinical diagnosis.

In this paper, we present a novel hybrid deformation model using global mass-spring medial representation structures and local finite element model. We employ the hybrid models, by fully calculating the FEM deformation in the local operation part while only calculating the global deformation by medial representation method. To achieve the real-time requirement of realistic deformable modeling, it is necessary to use the GPU parallel computing for FEM on regional deformation details, so the major calculation work in the conjugate gradient solver for the solution matrix is moved from CPU to GPU to accelerate the effectiveness. Evaluation and experiments are also discussed.

Software drift is a common phenomenon in software development processes, which may lead to process deviation and then affect software quality. Effectively controlling negative software drifts is the key to achieving acceptable, predictable, and dependable software evolution in the model-driven development. In this paper we put forward a general taxonomy to identify different drifts in development processes according to their effects on software development. Based on the taxonomy, categories of process drift and quality drift are detailedly introduced. Moreover, we propose an integration framework for different drifts to realize the integration between development process and product quality. Eventually, a case study from practical project development is shown to prove the validity of our identification framework.