Learning IoT in Edge: Deep Learning for the Internet of Things with Edge Computing

By He Li, Kaoru Ota, Mianxiong Dong. IEEE Network, 2018

1,411 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/00c4632d2d926acabc18574c9c5b870709ae9450

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Computer Vision and Pattern Recognition

Citations in the corpus

FetchSGD: Communication-Efficient Federated Learning with Sketching (2020, neutral): “First, communication-efﬁciency is a necessity when training on the edge (Li et al., 2018), since clients typically connect to the central aggregator over slow connections ( 1 Mbps) (Lee et al., 2010).”
Machine Learning at the Network Edge: A Survey (2019, methodology): “diﬀerentedgedevices. They used the Caﬀe framework and tested ten CNN tasks with diﬀerent networks, claiming their online algorithm candeployDLtaskstoedgedevicesandmaximizetheoveralldecisionperformance[43]. Osia et al. introduced a new CNN architecture for analyzing multimedia data from diﬀerent IoT ”
A Survey of Machine Learning Techniques Applied to Software Defined Networking (SDN): Research Issues and Challenges (2019, methodology): “In order to reduce network bandwidth consumption and improve the system response time, edge computing is often used to pre-process the raw data [257]–[259].”
A Survey of Multi-Access Edge Computing in 5G and Beyond: Fundamentals, Technology Integration, and State-of-the-Art (2019, neutral): “in [273] considered a two-layer DL model for video recognition with IoT devices.”
Privacy and Robustness in Federated Learning: Attacks and Defenses (2020, background): “Traditional centralized machine learning (ML) approaches cannot support such ubiquitous deployments and applications due to infrastructure shortcomings such as limited communication bandwidth, intermittent network connectivity, and strict delay constraints [1].”
Deep Learning in Mobile and Wireless Networking: A Survey (2018, background): “capability of devices [73], [378], [379], thus greatly benefiting users themselves, mobile operators, and indirectly device manufacturers.”
Thirty Years of Machine Learning: The Road to Pareto-Optimal Wireless Networks (2019, methodology): “[330] integrated both the DNN structure and the edge computing technique into the multimedia IoT, which was able to improve the efficiency of multimedia processing.”
A Review of Intrusion Detection Systems Using Machine and Deep Learning in Internet of Things: Challenges, Solutions and Future Directions (2020, background): “This enables automatic interactions between IoT-based systems in the absence of human intervention [171] to perform assigned collaborative functions.”
Data-Driven Methods for Predictive Maintenance of Industrial Equipment: A Survey (2019, background): “Specifically, model training is usually carried out on the cloud or server side because it requires strong computing power and storage space as support [31].”
Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing (2019, background): “Speciﬁcally, some frameworks [34]–[36] take advantage of DNN partitioning to optimize the computation ofﬂoading between the mobile devices and the cloud, while some framework target to distribute computation workload among mobile devices [37]–[39].”
Federated Learning Meets Blockchain in Edge Computing: Opportunities and Challenges (2021, background): “To accommodate the unpredictable demands of computation and storage for AI services, mobile edge computing (MEC) has been proposed for data training with low latency, by exploiting resources at the network edge near the data sources [3].”
Intelligent Edge Computing for IoT-Based Energy Management in Smart Cities (2019, background): “Second, EC is able to bridge the gap between the limited capability of low-powered devices and the computational demand of energy management [8].”
Communication-Efficient Federated Learning for Wireless Edge Intelligence in IoT (2020, neutral): “[20] put forward a similar idea where an ML model was trained in the cloud, then initial layers of the model were distributed to edge servers.”
Toward Massive Machine Type Communications in Ultra-Dense Cellular IoT Networks: Current Issues and Machine Learning-Assisted Solutions (2018, background): “Due to multi-layer structure of deep learning, it is considered as an effective approach for the edge-computing environment in order to accurately extract necessary information from the raw IoT sensor data [177].”
Edge Machine Learning for AI-Enabled IoT Devices: A Review (2020, methodology): “[45] divided the DNN model into two parts: The edge server computes the initial layers (lower layers) of the DNN model after it received the input data and then the cloud computes the higher layers of the DNN.”
Model Pruning Enables Efficient Federated Learning on Edge Devices (2019, background): “, 2019) and implementation on IoT devices (Li et al., 2018; Yao et al., 2017).”