Fog computing promises to enable machine learning tasks scale large amounts of data by distributing processing across connected devices. Two key challenges achieving this are (i) heterogeneity in devices' compute resources and (ii) topology constraints on which devices can communicate. We the first address these developing a network-aware distributed optimization methodology where process for task locally send their learnt parameters server aggregation at certain time intervals. Unlike...

Fog computing promises to enable machine learning tasks scale large amounts of data by distributing processing across connected devices. Two key challenges achieving this goal are (i) heterogeneity in devices' compute resources and (ii) topology constraints on which devices communicate with each other. We address these developing a novel network-aware distributed methodology where optimally share local send their learnt parameters server for periodic aggregation. Unlike traditional federated...

The emergence of the paradigm Internet Things (IoT) has necessitated development machine-to-machine (M2M) protocols geared towards wireless sensor network interfacing to and implementing machine learning algorithms over cloud. This paper discusses viability MQ Telemetry Transport (MQTT) protocol for such applications. introduces MQTT along with its merits demerits suitability IoT Then it outlines an implementation a typical application involving ubiquitous sensing, M2M communication, cloud...

Federated learning (FL) is a popular solution for distributed machine (ML). While FL has traditionally been studied supervised ML tasks, in many applications, it impractical to assume availability of labeled data across devices. To this end, we develop Cooperative unsupervised Contrastive Learning (CF-CL) facilitate edge devices with unlabeled datasets. CF-CL employs local device cooperation where either explicit (i.e., raw data) or implicit embeddings) information exchanged through...

Augmenting federated learning (FL) with direct device-to-device (D2D) communications can help improve conver-gence speed and reduce model bias through rapid local information exchange. However, data privacy concerns, device trust issues, unreliable wireless channels each pose challenges to determining an effective yet resource efficient D2D structure. In this paper, we develop a decentralized reinforcement (RL) methodology for graph discovery that promotes communication of non-sensitive...

The growing paradigm of Internet Things (IoT), has necessitated the development new and improved monitoring control methodologies for applications which were hitherto subjected to constraints a local system. This paper proposes predictive system monitor embedded linked IoT. A brief introduction its resultant effects on design are discussed followed by an overview regression systems. Next, we present prototype model implementing based model, inferences from data it generated execution....

Federated learning (FL) has been recognized as one of the most promising solutions for distributed machine (ML). In current literature, FL studied supervised ML tasks, in which edge devices collect labeled data. Nevertheless, many applications, it is impractical to assume existence data across devices. To this end, we develop a novel methodology, Cooperative unsupervised Contrastive Learning (CF-CL), with unlabeled datasets. CF-CL employs local device cooperation where are exchanged among...

In this paper, we present an algorithm based on reinforcement learning for scheduling the operations of industrial plant which is modeled as a network machines directed acyclic graph. The assumed to have access high-fidelity simulator plant, but not mathematical model. designed optimize objective function over moving window, similar receding horizon control, typical converts raw material into finished products. delivery schedule incoming be known subject uncertainty. A novel feature our...

One of the main challenges decentralized machine learning paradigms such as Federated Learning (FL) is presence local non-i.i.d. datasets. Device-to-device transfers (D2D) between distributed devices has been shown to be an effective tool for dealing with this problem and robust stragglers. In unsupervised case, however, it not obvious how data exchanges should take place due absence labels. paper, we propose approach create optimal graph transfer using Reinforcement Learning. The goal form...

Federated learning (FL) is a popular solution for distributed machine (ML). While FL has traditionally been studied supervised ML tasks, in many applications, it impractical to assume availability of labeled data across devices. To this end, we develop Cooperative unsupervised Contrastive Learning ({\tt CF-CL)} facilitate edge devices with unlabeled datasets. {\tt CF-CL} employs local device cooperation where either explicit (i.e., raw data) or implicit embeddings) information exchanged...

Fog computing promises to enable machine learning tasks scale large amounts of data by distributing processing across connected devices. Two key challenges achieving this goal are heterogeneity in devices compute resources and topology constraints on which can communicate with each other. We address these developing the first network-aware distributed optimization methodology where optimally share local send their learnt parameters a server for aggregation at certain time intervals. Unlike...

Wireless Sensor Networks (WSN) are gaining popularity as being the backbone of Cyber physical systems, IOT and various data acquisition from sensors deployed in remote, inaccessible terrains have remote deployment. However due to deployment, WSN is an adhoc network large number either heli-dropped terrain like volcanoes, Forests, border areas highly energy deficient available numbers. This makes it right soup become vulnerable kinds Security attacks. The lack resources deprived developing a...

Augmenting federated learning (FL) with direct device-to-device (D2D) communications can help improve convergence speed and reduce model bias through rapid local information exchange. However, data privacy concerns, device trust issues, unreliable wireless channels each pose challenges to determining an effective yet resource efficient D2D structure. In this paper, we develop a decentralized reinforcement (RL) methodology for graph discovery that promotes communication of non-sensitive...

Federated learning (FL) has been recognized as one of the most promising solutions for distributed machine (ML). In current literature, FL studied supervised ML tasks, in which edge devices collect labeled data. Nevertheless, many applications, it is impractical to assume existence data across devices. To this end, we develop a novel methodology, Cooperative unsupervised Contrastive Learning (CF-CL), with unlabeled datasets. CF-CL employs local device cooperation where are exchanged among...