The purpose of this paper is to contribute towards the near-future privacy-preserving big data analytical healthcare platforms, capable of processing streamed or uploaded timeseries data or videos from patients. The e...
详细信息
Matrix trace estimation is ubiquitous in machine learning applications and has traditionally relied on Hutchinson's method, which requires O(log(1/delta)epsilon(2)) matrix-vector product queries to achieve a (1 +/...
详细信息
ISBN:
(纸本)9781713845393
Matrix trace estimation is ubiquitous in machine learning applications and has traditionally relied on Hutchinson's method, which requires O(log(1/delta)epsilon(2)) matrix-vector product queries to achieve a (1 +/- epsilon)-multiplicative approximation to tr(A) with failure probability delta on positive-semidefinite input matrices A. Recently, the Hutch++ algorithm was proposed, which reduces the number of matrix-vector queries from O(1/epsilon(2)) to the optimal O(1/epsilon), and the algorithm succeeds with constant probability. However, in the high probability setting, the non-adaptive Hutch++ algorithm suffers an extra O (root log(1/delta)) multiplicative factor in its query complexity. Non-adaptive methods are important, as they correspond to sketching algorithms, which are mergeable, highly parallelizable, and provide low-memory streaming algorithms as well as low-communication distributed protocols. In this work, we close the gap between non-adaptive and adaptive algorithms, showing that even non-adaptive algorithms can achieve O(root log(1/delta)/epsilon + log(1/delta)) matrix-vector products. In addition, we prove matching lower bounds demonstrating that, up to a log log(1/delta) factor, no further improvement in the dependence on delta or epsilon is possible by any non-adaptive algorithm. Finally, our experiments demonstrate the superior performance of our sketch over the adaptive Hutch++ algorithm, which is less parallelizable, as well as over the non-adaptive Hutchinson's method.
In this work, we develop adaptive schemes using goal-oriented error control for a highly nonlinear flow temperature model with temperature dependent density. The dual-weighted residual method for computing error indic...
详细信息
Resilience has emerged as a crucial concept for evaluating structural performance under disasters because of its ability to extend beyond traditional risk assessments, accounting for a system's ability to minimize...
详细信息
An innovative adaptive strategy is applied to the super-twisting sliding mode observer (STSMO) for permanent magnet synchronous linear motors (PMSLM), addressing the issue of significant nonlinear variations in the ob...
详细信息
ISBN:
(数字)9798350368604
ISBN:
(纸本)9798350368611
An innovative adaptive strategy is applied to the super-twisting sliding mode observer (STSMO) for permanent magnet synchronous linear motors (PMSLM), addressing the issue of significant nonlinear variations in the observed back electromotive force (back-EMF) amplitude during uniform acceleration when using a fixed-parameter STSMO. The stability of the proposed method and the reasons behind the nonlinear behavior of the STSMO under fixed parameters are analyzed. A normalized orthogonal phase-locked loop (PLL) is employed to extract velocity and position. Simulation results demonstrate that the back-EMF observed by the new adaptive algorithm during uniform acceleration closely aligns with theoretical values, and the sensorless algorithm exhibits excellent tracking performance.
Bubbles are prevalent across key industries such as energy, chemical, biomedical, and oil and gas transportation, and have a significant impact on production efficiency and process safety. In this paper, an adaptive a...
详细信息
ISBN:
(数字)9798350380903
ISBN:
(纸本)9798350380910
Bubbles are prevalent across key industries such as energy, chemical, biomedical, and oil and gas transportation, and have a significant impact on production efficiency and process safety. In this paper, an adaptive algorithm for bubble identification utilizing laser scanning is developed in order to better extract bubbles with varying brightness intensity in the original image. The grayscale distributions of the rows and columns of the captured slice images are explored, where the presence of a bubble displays a distinct tendency to be first flat, then decreasing, next increasing, and finally flat. Therefore, the grayscale is divided into three parts: background, transition region, and bubble. The bubble edge can be determined by the grayscale distribution and gradient in the transition region. The flowchart of this adaptive algorithm is established. Based on the proposed algorithm, the identification of bubbles in a tank is achieved and the results are compared with those acquired by the Otsu method, revealing the better performance of the proposed algorithm. Besides, the bubble shape is examined, and the bubbles are visualized subsequently.
As interactions between humans and AI become more prevalent, it is critical to have better predictors of human behavior in these interactions. We investigated how changes in the AI’s adaptive algorithm impact behavio...
详细信息
Input distribution shift presents a significant problem in many real-world systems. Here we present Xenovert, an adaptive algorithm that can dynamically adapt to changes in input distribution. It is a perfect binary t...
详细信息
This paper proposes a predefined-time adaptive algorithm for the optimization problem with private inequality constraints. By designing algorithms, penalty parameters are designed to handle private inequality constrai...
详细信息
ISBN:
(数字)9798331508661
ISBN:
(纸本)9798331508678
This paper proposes a predefined-time adaptive algorithm for the optimization problem with private inequality constraints. By designing algorithms, penalty parameters are designed to handle private inequality constraints, and through adaptive penalties, the parameters in the algorithm are adaptively adjusted according to the degree of violation of private inequality constraints by the state solution. Moreover, The smoothing method is also integrated into adaptive penalty, overcoming the difficulty of non smooth dynamic behavior of the objective function and adaptive penalty parameters. In addition, a predefined-time mechanism is introduced to enable the algorithm to converge within the predefined-time.
Upcoming sensor networks would be deployed with sensing devices with energy harvesting capabilities from renewable energy sources such as solar power. A key research question in such sensor systems is to maximize the ...
详细信息
Upcoming sensor networks would be deployed with sensing devices with energy harvesting capabilities from renewable energy sources such as solar power. A key research question in such sensor systems is to maximize the asymptotic event detection probability achieved in the system, in the presence of energy constraints and uncertainties. This paper focuses on the design of adaptive algorithms for sensor activation in the presence of uncertainty in the event phenomena. Based upon the ideas from increase/decrease algorithms used in TCP congestion avoidance, we design an online and adaptive activation algorithm that varies the subsequent sleep interval according to additive increase and multiplicative decrease depending upon the sensor's current energy level. In addition, the proposed algorithm does not depend on global system parameters, or on the degree of event correlations, and hence can easily be deployed in practical scenarios. We analyze the performance of proposed algorithm for a single sensor scenario using Markov chains, and show that the proposed algorithm achieves near-optimal performance. Through extensive simulations, we demonstrate that the proposed algorithm not only achieves near-optimal performance, but also exhibits more stability with respect to sensor's energy level and sleep interval variations. We validate the applicability of our proposed algorithm in the presence of multiple sensors and multiple event processes through simulations. (c) 2011 Elsevier B.V. All rights reserved.
暂无评论