Richmond and Richmond (American Mathematical Monthly 104 (1997), 713-719) proved the following theorem: If, in a metric space with at least five points, all triangles are degenerate, then the space is isometric to a s...
详细信息
For individuals with diabetes, the intraperitoneal drug-delivery route may enable fully automated artificial pancreas technology. For such systems, the model predictive control (MPC) algorithm is favorable. However, M...
详细信息
ISBN:
(数字)9783907144107
ISBN:
(纸本)9798331540920
For individuals with diabetes, the intraperitoneal drug-delivery route may enable fully automated artificial pancreas technology. For such systems, the model predictive control (MPC) algorithm is favorable. However, MPC requires a reliable predictive model. In this work, we aim to design a trial protocol to collect data for identification of a bi-hormonal intraperitoneal prediction model. We apply model-based design of experiment (MBDoE) to determine the optimal input of meals, subcutaneous insulin injections, and subcutaneous glucagon injections. Based on parameters from two anesthetized pigs, we design experiments to identify parameters in awake animals. Our results demonstrate how MBDoE may be used as a planning tool when designing trial protocols. The approach may hold potential as a support tool for clinicians when personalizing control algorithms for human AP users.
This study applies the Fractional Reduced Differential Transform Method (FRDTM) to solve two nonlinear fractional equations: the time-fractional Schrödinger equation (TFSE) and the coupled Schrödinger–KdV (...
详细信息
Despite the recent success of artificial neural networks on a variety of tasks, we have little knowledge or control over the exact solutions these models implement. Instilling inductive biases — preferences for some ...
详细信息
This paper presents a parallel Monte Carlo simulation based performance quantification method for nonlinear model predictive control (NMPC) in closed-loop. The method provides distributions for the controller performa...
This paper presents a parallel Monte Carlo simulation based performance quantification method for nonlinear model predictive control (NMPC) in closed-loop. The method provides distributions for the controller performance in stochastic systems enabling performance quantification. We perform high-performance Monte Carlo simulations in C enabled by a new thread-safe NMPC implementation in combination with an existing high-performance Monte Carlo simulation toolbox in C. We express the NMPC regulator as an optimal control problem (OCP), which we solve with the new thread-safe sequential quadratic programming software NLPSQP. Our results show almost linear scale-up for the NMPC closed-loop on a 32 core CPU. In particular, we get approximately 27 times speed-up on 32 cores. We demonstrate the performance quantification method on a simple continuous stirred tank reactor (CSTR), where we perform 30,000 closed-loop simulations with both an NMPC and a reference proportional-integral (PI) controller. Performance quantification of the stochastic closed-loop system shows that the NMPC outperforms the PI controller in both mean and variance.
Due to the disparity in the levels of difficulty presented by the several tasks, doing domain adaptation in an adversarial way may result in an imbalanced learning process. In the MNIST dataset, this phenomenon also m...
Due to the disparity in the levels of difficulty presented by the several tasks, doing domain adaptation in an adversarial way may result in an imbalanced learning process. In the MNIST dataset, this phenomenon also manifests itself in the form of domain adaptation for color-shifted distribution. In this particular situation, the domain classifier has a higher tendency to fit more quickly, but the category classifier fits quite poorly in the learning process. In order to address this problem, a new hyper-parameter has been added to the loss function in order to strike a compromise between the learning speed of the domain and the categorical classifier. By using this technique, the categorical classifier may better match the data while still maintaining the same level of performance as the domain classifier. In order to determine whether or not making use of this hyper-parameter is useful, the phenomena in question is examined using three distinct color-shifted settings. Following the evaluations, it was discovered that the newly introduced hyper-parameter is capable of coping with imbalanced learning while simultaneously engaging in domain adaptation.
Conduction graphs are defined here in order to elucidate at a glance the often complicated conduction behaviour of molecular graphs as ballistic molecular conductors. The graph GC describes all possible conducting dev...
详细信息
The paper presents an overview of the third edition of the shared task on multilingual coreference resolution, held as part of the CRAC 2024 workshop. Similarly to the previous two editions, the participants were chal...
Smart grid management is an emerging research topic that recently has adopted artificial intelligence algorithms to assist in the task. However, as more and more data is used, data insecurity and cyber-physical attack...
Smart grid management is an emerging research topic that recently has adopted artificial intelligence algorithms to assist in the task. However, as more and more data is used, data insecurity and cyber-physical attacks hinder the performance of intelligent systems. In this paper, we propose a fuzzy electricity management system (FEMS) consisting of an attention-based anomaly detection module for attack classification and a fuzzy Q-learning decision module for grid management. Experimental results show that the proposed anomaly detection module achieves high accuracy and F1 scores, significantly outperforming state-of-the-art systems. As for the management evaluation, FEMS achieves extremely low convergence days and mean absolute error (MAE) of supply distribution, which proves the effectiveness of the proposed FEMS in shaping supply distribution. Moreover, FEMS achieves the lowest failure rate (highest stability) but a slightly higher MAE of operating reserve rate due to the unavoidable trade-off between grid stability and energy efficiency.
Graph representation learning (GRL) has become a prominent tool for furthering the understanding of complex networks providing tools for network embedding, link prediction, and node classification. In this paper, we p...
详细信息
暂无评论