Squeezing Deep Learning into Mobile and Embedded Devices

Lane, Nicholas D.; Bhattacharya, Sourav; Mathur, Akhil; Georgiev, Petko; Forlivesi, Claudio; Kawsar, Fahim

doi:10.1109/mprv.2017.2940968

Cited by 174 publications

(72 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For example, the paper considered a simple enough NN model for optimization running on IoT devices with limited computation capabilities. However, training of deeper networks on constrained devices is expected to become the mainstream in the near future [48]. Application of consensus based federated optimization to deeper networks might require a more efficient use of the limited bandwidth, including quantization, compression or ad-hoc channel encoding [47].…”

Section: Conclusion and Open Problemsmentioning

confidence: 99%

Federated Learning With Cooperating Devices: A Consensus Approach for Massive IoT Networks

Savazzi¹,

Nicoli²,

Rampa³

2020

IEEE Internet Things J.

318

164

View full text Add to dashboard Cite

Federated learning (FL) is emerging as a new paradigm to train machine learning models in distributed systems. Rather than sharing, and disclosing, the training dataset with the server, the model parameters (e.g., neural networks weights and biases) are optimized collectively by large populations of interconnected devices, acting as local learners. FL can be applied to power-constrained IoT devices with slow and sporadic connections. In addition, it does not need data to be exported to third parties, preserving privacy. Despite these benefits, a main limit of existing approaches is the centralized optimization which relies on a server for aggregation and fusion of local parameters; this has the drawback of a single point of failure and scaling issues for increasing network size. The paper proposes a fully distributed (or server-less) learning approach: the proposed FL algorithms leverage the cooperation of devices that perform data operations inside the network by iterating local computations and mutual interactions via consensus-based methods. The approach lays the groundwork for integration of FL within 5G and beyond networks characterized by decentralized connectivity and computing, with intelligence distributed over the end-devices. The proposed methodology is verified by experimental datasets collected inside an industrial IoT environment.

show abstract

Section: Conclusion and Open Problemsmentioning

confidence: 99%

Federated Learning With Cooperating Devices: A Consensus Approach for Massive IoT Networks

Savazzi¹,

Nicoli²,

Rampa³

2020

IEEE Internet Things J.

318

164

View full text Add to dashboard Cite

show abstract

“…Moreover, capabilities to develop Deep Learning algorithms could be added. They are well suited for classification problems, but deploying them on a constrained device is challenging [16].…”

Section: Limitations and Perspectivesmentioning

confidence: 99%

Model driven programming of autonomous floats for multidisciplinary monitoring of the oceans

Bonnieux

Mosser

Blay–Fornarino

et al. 2019

OCEANS 2019 - Marseille

View full text Add to dashboard Cite

Monitoring of the oceans with autonomous floats is of great interest for many disciplines. Monitoring on a global scale needs a multidisciplinary approach to be affordable. For this purpose, we propose an approach that allows oceanographers from different specialities to develop applications for autonomous floats. However, developing such applications usually requires expertise in embedded systems, and they must be reliable and efficient with regards to the limited resources of the floats (e.g., energy, processing power). We have followed a Model Driven Engineering approach composed of i) a Domain Specific Language to allow oceanographers to develop applications, ii) analysis tools to ensure that applications are efficient and reliable, iii) a composition tool to allow the deployment of different applications on a same float, and iv) a code generator that produce efficient and reliable code for the float. We present our approach with a biological and a seismological application. We validate it with technical metrics and an experiment.

show abstract

“…Feasibility & Performance: Recent hardware advancements allow to embed a fair amount of computational power in sensor and actuator devices as well as in single chips [36]. In parallel, machine learning techniques are also steadily improving, so that incorporating "intelligence" within smart objects is already a feasible reality [16], [7].…”

Section: Challenges and Benefitsmentioning

confidence: 99%

An Argumentation-Based Perspective Over the Social IoT

Lippi

Mamei

Mariani

et al. 2018

IEEE Internet Things J.

View full text Add to dashboard Cite

The crucial role played by social interactions between smart objects in the Internet of Things is being rapidly recognized by the Social Internet of Things (SIoT) vision. In this paper, we build upon the recently introduced vision of Speaking Objects -"things" interacting through argumentation -to show how different forms of human dialogue naturally fit cooperation and coordination requirements of the SIoT. In particular, we show how speaking objects can exchange arguments in order to seek for information, negotiate over an issue, persuade others, deliberate actions, and so on, namely, striving to reach consensus about the state of affairs and their goals. In this context, we illustrate how argumentation naturally enables such a form of conversational coordination through practical examples and a case study scenario.

show abstract

Squeezing Deep Learning into Mobile and Embedded Devices

Cited by 174 publications

References 7 publications

Federated Learning With Cooperating Devices: A Consensus Approach for Massive IoT Networks

Federated Learning With Cooperating Devices: A Consensus Approach for Massive IoT Networks

Model driven programming of autonomous floats for multidisciplinary monitoring of the oceans

An Argumentation-Based Perspective Over the Social IoT

Contact Info

Product

Resources

About