Accurate blood pressure (BP) monitoring using photoplethysmography (PPG) remains challenging due to noise, individual variability, and nonlinear signal dynamics. In this work, we present FBPP-Net, a lightweight temporal-mixing framework that integrates sparse and shared Mixture-of-Experts (MoE) modules with quantile regression. The model enables specialized subnetworks to capture diverse temporal dependencies while providing robust probabilistic estimation of systolic and diastolic BP. Without requiring GPU acceleration, FBPP-Net achieves efficient training and inference, making it suitable for real-time wearable applications. Experiments on the UQVS dataset show that FBPP-Net-MoE attains SBP/DBP errors of 2.83/3.54 mmHg, and FBPP-Net-TM achieves 3.08/3.18 mmHg, outperforming XGBoost, LSTM, MLP, Informer, and TSMixer baselines. Furthermore, the analysis of expert activations and temporal segments provides interpretable insights into the localized dynamics driving near-future BP variations, supporting intelligent and explainable physiological monitoring.

In an integrated energy system (IES), fluctuations in coupled district heating networks and renewable energy sources pose risks to the power system’s operation. To effectively mitigate the risk of rare events, it is crucial to identify the key factors that influence their likelihood. Traditional global sensitivity analysis (GSA) assesses the sensitivity of full probability density functions (PDFs) of the system output to uncertain inputs. However, it fails to evaluate the sensitivity of rare event probabilities, which are at the tail of the PDF. It also neglects the uncertainties of the input PDFs, whose hyperparameters can heavily affect the rare event probabilities. To address these issues, we propose a novel framework calledrare-event-based global sensitivity analysis(REGSA) that prioritizes the input PDF parameters that impact the risks of the operating state. We also improve the computational efficiency of this framework by quantifying rare event probabilities through subset simulation and using sparse polynomial chaos expansion (SPCE) in REGSA (hereafter referred to as SPCE-REGSA). The simulations reveal the excellent performance of the proposed method.

We present a novel method for graph partitioning, based on reinforcement learning and graph convolutional neural networks. Our approach is to recursively partition coarser representations of a given graph. The neural network is implemented using SAGE graph convolution layers, and trained using an advantage actor critic (A2C) agent. We present two variants, one for finding ean edge separator that minimizes the normalized cut or quotient cut, and one that finds a small vertex separator. The vertex separators are then used to construct a nested dissection ordering to permute a sparse matrix so that its triangular factorization will incur less fill-in. The partitioning quality is compared with partitions obtained using METIS and SCOTCH, and the nested dissection ordering is evaluated in the sparse solver SuperLU. Our results show that the proposed method achieves similar partitioning quality as METIS, SCOTCH and spectral partitioning. Furthermore, the method generalizes across different classes of graphs, and works well on a variety of graphs from the SuiteSparse sparse matrix collection.