The way to provide energy supply for electric passenger vehicles by adopting the power exchange mode is an important guarantee to promote the development of the Electric Passenger Vehicles (EPVs) industry. This paper ...
详细信息
The Generative Artificial Intelligence chatbots (GAI chatbots) have emerged as promising tools in various domains, including higher education, so this study aims to investigate the role of Bard, a newly developed GAI ...
详细信息
Existing algorithms for predicting milling chatter have not been widely adopted in industry since they require specialized instruments to measure the stability inputs. This study describes how the machining process fo...
详细信息
The utilization of 3D scanning and reverse engineering techniques has revolutionized quality control practices across various industries. These technologies play a pivotal role in accurately assessing and documenting ...
详细信息
In order to solve the problems of lack of data support, excessive subjectivity of expert judgment, fuzzy semantics, and mutual coupling between indices in safety evaluation process of tactical training airspace planni...
详细信息
Safety is a critical concern for the next generation of autonomy that is likely to rely heavily on deep neural networks for perception and control. This paper proposes a method to repair unsafe ReLU DNNs in safety-cri...
详细信息
ISBN:
(纸本)9783031158391;9783031158384
Safety is a critical concern for the next generation of autonomy that is likely to rely heavily on deep neural networks for perception and control. This paper proposes a method to repair unsafe ReLU DNNs in safety-critical systems using reachability analysis. Our repair method uses reachability analysis to calculate the unsafe reachable domain of a DNN, and then uses a novel loss function to construct its distance to the safe domain during the retraining process. Since subtle changes of the DNN parameters can cause unexpected performance degradation, we also present a minimal repair approach where the DNN deviation is minimized. Furthermore, we explore applications of our method to repair DNN agents in deep reinforcement learning (DRL) with seamless integration with learning algorithms. Our method is evaluated on the ACAS Xu benchmark and a rocket lander system against the state-of-the-artmethod ART. Experimental results show that our repair approach can generate provably safe DNNs on multiple safety specifications with negligible performance degradation, even in the absence of training data (Code is available online at https://***/Shaddadi/***).
Digital networked control systems are of growing importance in safety-critical systems and perform indispensable function in most complex systems today. Networked degradation such as transmission delay cause such syst...
详细信息
This paper proposes a new Gradient Boosting Component-decision (GBC-D) model, which combines the advantages of Markov Decision process (MDP), gradient lifting tree (GBT) and principal Component analysis (PCA). Applied...
详细信息
ISBN:
(数字)9798350389579
ISBN:
(纸本)9798350389586
This paper proposes a new Gradient Boosting Component-decision (GBC-D) model, which combines the advantages of Markov Decision process (MDP), gradient lifting tree (GBT) and principal Component analysis (PCA). Applied to big dataprocessing is based on machine learning in market trend prediction model research. The GBC-D model uses PCA for data reduction and feature extraction, which reduces the computation and preserves the main features of the data. MDP is used to describe the market state transition process and maximize the expected cumulative return by optimizing the decision sequence. GBT is introduced as a forecasting model to gradually improve the forecasting performance to predict market trends more accurately. Combining the advantages of the three methods, the model can effectively deal with the complexity and uncertainty in big data and improve the accuracy and stability of market trend prediction. In the experimental part, we compared the BGC-D with the traditional prediction model. The results show that GBC-D can reduce the prediction error in the complex and changeable market environment, capture the market dynamics more accurately, and improve the overall performance of the system. This fully validates the feasibility and effectiveness of GBC-D architecture in big dataprocessing market trend prediction.
This article discusses methods for free-form surface quality and the amplitude parameters of roughness, based on research and field experiments to control the condition of the tool. The methods of correlation analysis...
详细信息
This article presents the development of an intelligent system for classifying wrist extension and flexion movements using electromyography (EMG). The system utilizes an EMG signal placed on the subject's upper li...
详细信息
暂无评论