98%
921
2 minutes
20
The Fokker-Planck (FP) equation governs the probabilistic response of diffusion processes driven by stochastic differential equations (SDEs). Gaussian mixture models and deep learning solvers are two state-of-the-art methods for solving the FP equation. Although mixture models mostly depend on empirical sampling strategies and predefined Gaussian components, deep learning techniques suffer from inherent interpretability deficits and require excessively large training samples. To address these challenges, we propose an adaptive normalizing flow framework for solving FP equations (ANFFP). Normalizing flows are generative models that produce tractable distributions to approximate the complex target distributions. The ANFFP architecture inherently preserves probabilistic interpretability while enabling efficient exact sampling advantages that significantly enhance its applicability to probabilistic response modeling under small sample conditions. Numerical examples involving one-dimensional, two-dimensional, and four-dimensional SDEs demonstrate the effectiveness of the method. In addition, the computational complexity of the ANFFP method is discussed in more detail. This work provides a new paradigm for solving high-dimensional FP equations with theoretical guarantees and practical scalability.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1063/5.0273776 | DOI Listing |
Neural Netw
September 2025
School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, Shanghai, 200240, China; Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong University, Shanghai, 200240, China; Institute of Medical Robotics, Shanghai Jiao Tong University, Shanghai, 200240, China.
3D shape defect detection plays an important role in autonomous industrial inspection. However, accurate detection of anomalies remains challenging due to the complexity of multimodal sensor data, especially when both color and structural information are required. In this work, we propose a lightweight inter-modality feature prediction framework that effectively utilizes multimodal fused features from the inputs of RGB, depth and point clouds for efficient 3D shape defect detection.
View Article and Find Full Text PDFJ Am Soc Mass Spectrom
September 2025
Anhui Province Key Laboratory for Control and Applications of Optoelectronic Information Materials, School of Physics and Electronic Information, Anhui Normal University, Wuhu, Anhui 241000, China.
An integrated miniature time-of-flight mass spectrometer (TOF-MS) system coupled with a pocket-size 3D-printed laser-induced acoustic desorption (LIAD) source is described. This 3D-printed LIAD source utilizes only a miniature deceleration motor to achieve two-dimensional motion of the target surface, simplifying the source structure and improving the long-term stability of mass spectrometry measurements. It has been successfully applied to analyze the model molecule creatinine and ingredients in an energy beverage (Red Bull), where main natural nutrients were clearly identified.
View Article and Find Full Text PDFPLoS One
September 2025
School of Computer Science and Engineering, Southeast University, China.
Metaheuristic optimization algorithms often face challenges such as complex modeling, limited adaptability, and a tendency to get trapped in local optima when solving complex optimization problems. To enhance algorithm performance, this paper proposes an enhanced Secretary Bird Optimization Algorithm (MESBOA) based on a precise elimination mechanism and boundary control. The algorithm integrates three key strategies: a precise population elimination strategy, which optimizes the population structure by eliminating individuals with low fitness and intelligently generating new ones; a lens imaging-based opposition learning strategy, which expands the exploration of the solution space through reflection and scaling to reduce the risk of local optima; and a boundary control strategy based on the best individual, which effectively constrains the search range to avoid inefficient searches and premature convergence.
View Article and Find Full Text PDFACS Nano
September 2025
Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon 16419, Republic of Korea.
Vagus nerve stimulation (VNS) is a promising therapy for neurological and inflammatory disorders across multiple organ systems. However, conventional rigid interfaces fail to accommodate dynamic mechanical environments, leading to mechanical mismatches, tissue irritation, and unstable long-term interfaces. Although soft neural interfaces address these limitations, maintaining mechanical durability and stable electrical performance remains challenging.
View Article and Find Full Text PDFInt J Biometeorol
September 2025
Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, 100101, China.
Plant viewing activities, which encompass the enjoyment of seasonal plant phenomena such as flowering and autumn leaf coloration, have become popular worldwide. Plant viewing activities are increasingly challenged by climate change, as key components like plant phenology and climate comfort are highly sensitive to global warming. However, few studies have explored the impact of climate change on viewing activities, particularly from an integrated, multi-factor perspective.
View Article and Find Full Text PDF