An Internet of Agents Architecture for Training and Deployment of Deep Convolutional Models
- PDF / 1,051,267 Bytes
- 9 Pages / 595.224 x 790.955 pts Page_size
- 106 Downloads / 191 Views
An Internet of Agents Architecture for Training and Deployment of Deep Convolutional Models Luis Rodriguez-Benitez1
1 · Ramon Hervas1 · L. Jimenez-Linares1 ˜ · Carlos Cordoba-Ruiz1 · Luis Cabanero
Received: 1 November 2019 / Revised: 27 August 2020 / Accepted: 28 September 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract It is a fact that Artificial Intelligence is having an ever-growing impact on society. That is not just because of advances in computational power and in machine learning models, such as deep neural networks, but also because of the availability of a large volume of heterogeneous data from diverse sources. The Internet of Things (IoT) paradigm is helping gather massive amounts of data from sensor networks that can be used to train and generate complex AI models. However, the training of these models needs not only the data but has high computational requirements. In this scenario there has appeared a new paradigm, called the Internet of Agents (IoA), which allows the inclusion of intelligence and autonomy in IoT devices and networks. This paper presents an IoA architecture that allows the continual and distributed generation and exploitation of convolutional neural networks. Specific protocols for the safe and efficient transmission of models and training pictures are designed. The convolutional model is trained in the cloud and, once reduced, it is distributed and executed in agents located in embedded devices with low computational resources. The architecture has been tested using a convolutional model for the recognition of handwritten character digits based on the MNIST database. Keywords Internet of agents · Embedded systems · Convolutional neural networks
1 Introduction Nowadays, devices in the IoT are expected to interact with each other and with their environment, exchanging the information received from diverse sensors. If a direct correspondence between the operation of an IoT node and reactive agents is established, it can be stated that such operation is very limited due to the fact that they only take into account data from the current state of the environment and they usually react instantaneously without using a complex reasoning process. Anyway, this simplicity is not valid for more complex environments where information from previous states of the environments is needed and so is a high level reasoning process.
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11265-020-01602-6) contains supplementary material, which is available to authorized users. Luis Rodriguez-Benitez
[email protected] 1
Escuela Superior de Informatica, Universidad de Castilla-La Mancha, 13071, Ciudad Real, Spain
From intelligent agents in complex environments to a multi-agent system (MAS), research on intelligent agents has grown over the last few decades. Mainly, agent technology has been employed over the last decade to specifically access, gather, and integrate information from different web sources. Currently, th
Data Loading...
