Machine learning, a vital part of artificial intelli-gence, improves our ability to make predictions from complex data. The success of these predictions relies heavily on the model's fit with its data and the data...
详细信息
ISBN:
(数字)9798350378078
ISBN:
(纸本)9798350378085
Machine learning, a vital part of artificial intelli-gence, improves our ability to make predictions from complex data. The success of these predictions relies heavily on the model's fit with its data and the data's consistency. Accurate predictions depend on closely matching the model to the data patterns. However, missing data can seriously disrupt the accuracy of predictions, causing errors. To address this, we recommend adopting a proposed framework that enhances predictions by effectively filling in missing data. Our objective is to refine the accuracy of our predictions, especially in scenarios where the data might be incomplete and features exhibit patterns of dependency. In the context of missing data, our approach concentrates on enhancing regression performance to minimize the effects of missing data as effectively as possible.
Indoor localization has gained significant attention in recent years due to its various applications in smart homes, industrial automation, and healthcare, especially since more people rely on their wireless devices f...
Indoor localization has gained significant attention in recent years due to its various applications in smart homes, industrial automation, and healthcare, especially since more people rely on their wireless devices for location-based services. Deep learning-based solutions have shown promising results in accurately estimating the position of wireless devices in indoor environments using wireless parameters such as Channel State Information (CSI) and Received Signal Strength Indicator (RSSI). However, despite the success of deep learning-based approaches in achieving high localization accuracy, these models suffer from a lack of generalizability and can not be readily-deployed to new environments or operate in dynamic environments without retraining. In this paper, we propose meta-learning-based localization models to address the lack of generalizability that persists in conventionally trained DL-based localization models. Furthermore, since meta-learning algorithms require diverse datasets from several different scenarios, which can be hard to collect in the context of localization, we design and propose a new meta-learning algorithm, TB-MAML (Task Biased Model Agnostic Meta Learning), intended to further improve generalizability when the dataset is limited. Lastly, we evaluate the performance of TB-MAML-based localization against conventionally trained localization models and localization done using other meta-learnina algorithms.
Analysis of a special machine topology, which features a doubly salient structure and permanent magnets in the stator and provides very high power density performance has been presented in this paper. An analytical mo...
详细信息
Direct policy search has been widely applied in modern reinforcement learning and continuous control. However, the theoretical properties of direct policy search on nonsmooth robust control synthesis have not been ful...
ISBN:
(纸本)9781713871088
Direct policy search has been widely applied in modern reinforcement learning and continuous control. However, the theoretical properties of direct policy search on nonsmooth robust control synthesis have not been fully understood. The optimal H∞ control framework aims at designing a policy to minimize the closed-loop H∞ norm, and is arguably the most fundamental robust control paradigm. In this work, we show that direct policy search is guaranteed to find the global solution of the robust H∞ state-feedback control design problem. Notice that policy search for optimal H∞ control leads to a constrained nonconvex nonsmooth optimization problem, where the nonconvex feasible set consists of all the policies stabilizing the closed-loop dynamics. We show that for this nonsmooth optimization problem, all Clarke stationary points are global minimum. Next, we identify the coerciveness of the closed-loop H∞ objective function, and prove that all the sublevel sets of the resultant policy search problem are compact. Based on these properties, we show that Goldstein's subgradient method and its implementable variants can be guaranteed to stay in the nonconvex feasible set and eventually find the global optimal solution of the H∞ state-feedback synthesis problem. Our work builds a new connection between nonconvex nonsmooth optimization theory and robust control, leading to an interesting global convergence result for direct policy search on optimal H∞ synthesis.
Recent advancements in deep learning have improved the signal recognition capabilities of distributed fiber-optic vibration sensors. However, most existing research focuses on closed-set scenarios, assuming that all p...
详细信息
As a new distributed machine learning methodology, Federated Learning (FL) allows mobile devices (MDs) to collaboratively train a global model without sharing their raw data in a privacy-preserving manner. However, it...
详细信息
ISBN:
(数字)9798350384475
ISBN:
(纸本)9798350384482
As a new distributed machine learning methodology, Federated Learning (FL) allows mobile devices (MDs) to collaboratively train a global model without sharing their raw data in a privacy-preserving manner. However, it is a great challenge to schedule each MD and allocate various resources reasonably. This paper studies the joint optimization of computing resources used by MDs for FL training, the number of local iterations as well as WPT duration of each MD in a Wireless Power Transfer (WPT) assisted FL system, with the goal of maximizing the total utility of all MDs in the entire FL training process. Furthermore, we analyze the problem by using the Karush-Kuhn-Tucker (KKT) conditions and Lagrange dual method, and propose an improved Lagrangian subgradient method to solve this problem. Finally, extensive simulation experiments are conducted under various scenarios to verify the effectiveness of the proposed algorithm. The results show that our proposed algorithm has better performance in terms of the total utility of all MDs compared with other benchmark methods.
In Software Defined Networking (SDN) centralization of network control creates a single point of failure and a valuable target for threat actors that wish to produce an impact on the network. Notably, network controll...
In Software Defined Networking (SDN) centralization of network control creates a single point of failure and a valuable target for threat actors that wish to produce an impact on the network. Notably, network controllers can be targeted by effects such as Denial of Service (DoS) attacks that would cripple the performance of the network as a whole by making critical services unavailable if they are not detected and prevented. In Artificial Intelligence (AI), Machine Learning (ML) techniques are used to identify the presence of malicious distributed DoS activity within a sample of data collected from network activity over an interval of time. However, machine learning techniques are themselves vulnerable to attacks. We perform a detailed security analysis of a highly realistic threat model and experimentally demonstrate the effectiveness of poisoning and evasion attacks on the SDN. After this adversarial AI experiment, we propose a defense mechanism adapted to the machine learning algorithm and able to protect against those attacks on the system. This research highlights the implications of the poor management of the use of AI as a new technology in this field.
Deep learning performance may decrease substantially with unseen heterogeneous data. While most unsupervised domain adaptation (UDA) methods seek to address this through image alignment, they often ignore uncertainty ...
详细信息
This study investigates the global consensus problem of mixed-order multi-agent systems. Firstly, a control strategy based on switching functions is proposed to ensure the effective operation of fuzzy control in the f...
详细信息
This study investigates the global consensus problem of mixed-order multi-agent systems. Firstly, a control strategy based on switching functions is proposed to ensure the effective operation of fuzzy control in the fuzzy region. Secondly, the differential equation model satisfying the global Lipschitz condition is extended, and a more universal model framework is constructed. Furthermore, online learning laws with time-varying σ-modifications are introduced to effectively guarantee the global asymptotic convergence of the closed-loop multi-agent systems. Finally, the effectiveness of the proposed algorithm is validated through designed simulation experiments.
暂无评论