scholarly journals Embedded Intelligence on FPGA: Survey, Applications and Challenges

Electronics ◽  
2021 ◽  
Vol 10 (8) ◽  
pp. 895
Author(s):  
Kah Phooi Seng ◽  
Paik Jen Lee ◽  
Li Minn Ang
Keyword(s):  
Machine Learning ◽  
Indoor Localization ◽  
High Energy ◽  
Research Field ◽  
Machine Learning Algorithms ◽  
Self Organizing Map ◽  
Intelligent Decision Making ◽  
Intelligent Decision ◽  
Field Programmable ◽  
Embedded Intelligence

Embedded intelligence (EI) is an emerging research field and has the objective to incorporate machine learning algorithms and intelligent decision-making capabilities into mobile and embedded devices or systems. There are several challenges to be addressed to realize efficient EI implementations in hardware such as the need for: (1) high computational processing; (2) low power consumption (or high energy efficiency); and (3) scalability to accommodate different network sizes and topologies. In recent years, an emerging hardware technology which has demonstrated strong potential and capabilities for EI implementations is the FPGA (field programmable gate array) technology. This paper presents an overview and review of embedded intelligence on FPGA with a focus on applications, platforms and challenges. There are four main classification and thematic descriptors which are reviewed and discussed in this paper for EI: (1) EI techniques including machine learning and neural networks, deep learning, expert systems, fuzzy intelligence, swarm intelligence, self-organizing map (SOM) and extreme learning; (2) applications for EI including object detection and recognition, indoor localization and surveillance monitoring, and other EI applications; (3) hardware and platforms for EI; and (4) challenges for EI. The paper aims to introduce interested researchers to this area and motivate the development of practical FPGA solutions for EI deployment.

HAND GESTURE RECOGNITION IN REAL-TIME

2021 ◽  
Vol 1 (10) ◽  
pp. 346-350
Author(s):  
Priyanshi Gupta ◽  
Amita Goel ◽  
Nidhi Sengar ◽  
Vashudha Bahl
Keyword(s):  
Machine Learning ◽  
Gesture Recognition ◽  
Research Field ◽  
Hand Gesture Recognition ◽  
Machine Learning Algorithms ◽  
Hand Gesture ◽  
Normal People ◽  
Human Gestures ◽  
Dynamic Gestures ◽  
Hand Pose

Hand gesture is language through which normal people can communicate with deaf and dumb people. Hand gesture recognition detects the hand pose and converts it to the corresponding alphabet or sentence. In past years it received great attention from society because of its application. It uses machine learning algorithms. Hand gesture recognition is a great application of human computer interaction. An emerging research field that is based on human centered computing aims to understand human gestures and integrate users and their social context with computer systems. One of the unique and challenging applications in this framework is to collect information about human dynamic gestures. Keywords: Tensor Flow, Machine learning, React js, handmark model, media pipeline

scholarly journals Unfolding Algorithms

2019 ◽  
Vol 32 (4) ◽  
pp. 119-136
Author(s):  
Bilel Benbouzid
Keyword(s):  
Machine Learning ◽  
Research Field ◽  
Machine Learning Algorithms ◽  
Moral Vision ◽  
Predictive Policing ◽  
Method Of Calculation ◽  
Predictive Algorithms ◽  
Cautious Interpretation ◽  
The University ◽  
Two Sides

Predictive policing is a research field whose principal aim is to develop machines for predicting crimes, drawing on machine learning algorithms and the growing availability of a diversity of data. This paper deals with the case of the algorithm of PredPol, the best-known startup in predictive policing. The mathematicians behind it took their inspiration from an algorithm created by a French seismologist, a professor in earth sciences at the University of Savoie. As the source code of the PredPol platform is kept inaccessible as a trade secret, the author contacted the seismologist directly in order to try to understand the predictions of the company’s algorithm. Using the same method of calculation on the same data, the seismologist arrived at a different, more cautious interpretation of the algorithm's capacity to predict crime. How were these predictive analyses formed on the two sides of the Atlantic? How do predictive algorithms come to exist differently in these different contexts? How and why is it that predictive machines can foretell a crime that is yet to be committed in a California laboratory, and yet no longer work in another laboratory in Chambéry?  In answering these questions, I found that machine learning researchers have a moral vision of their own activity that can be understood by analyzing the values and material consequences involved in the evaluation tests that are used to create the predictions.

Machine Learning Techniques for IoT-Based Indoor Tracking and Localization

2022 ◽  
pp. 123-145
Author(s):  
Pelin Yildirim Taser ◽  
Vahid Khalilpour Akram
Keyword(s):  
Machine Learning ◽  
Indoor Localization ◽  
Learning Algorithms ◽  
Distance Estimation ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
The Internet ◽  
Traditional Methods ◽  
Learning Techniques ◽  
Localization Systems

The GPS signals are not available inside the buildings; hence, indoor localization systems rely on indoor technologies such as Bluetooth, WiFi, and RFID. These signals are used for estimating the distance between a target and available reference points. By combining the estimated distances, the location of the target nodes is determined. The wide spreading of the internet and the exponential increase in small hardware diversity allow the creation of the internet of things (IoT)-based indoor localization systems. This chapter reviews the traditional and machine learning-based methods for IoT-based positioning systems. The traditional methods include various distance estimation and localization approaches; however, these approaches have some limitations. Because of the high prediction performance, machine learning algorithms are used for indoor localization problems in recent years. The chapter focuses on presenting an overview of the application of machine learning algorithms in indoor localization problems where the traditional methods remain incapable.

scholarly journals An Accelerator Architecture of Changeable-Dimension Matrix Computing Method for SVM

Electronics ◽  
2019 ◽  
Vol 8 (2) ◽  
pp. 143 ◽  
Author(s):  
Ruidong Wu ◽  
Bing Liu ◽  
Ping Fu ◽  
Junbao Li ◽  
Shou Feng
Keyword(s):  
Machine Learning ◽  
Learning Algorithms ◽  
Matrix Multiplication ◽  
Critical Time ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Practical Applications ◽  
The Matrix ◽  
Field Programmable ◽  
Computing Method

Matrix multiplication is a critical time-consuming processing step in many machine learning applications. Due to the diversity of practical applications, the matrix dimensions are generally not fixed. However, most matrix calculation methods, based on field programmable gate array (FPGA) currently use fixed matrix dimensions, which limit the flexibility of machine learning algorithms in a FPGA. The bottleneck lies in the limited FPGA resources. Therefore, this paper proposes an accelerator architecture for matrix computing method with changeable dimensions. Multi-matrix synchronous calculation concept allows matrix data to be processed continuously, which improves the parallel computing characteristics of FPGA and optimizes the computational efficiency. This paper tests matrix multiplication using support vector machine (SVM) algorithm to verify the performance of proposed architecture on the ZYNQ platform. The experimental results show that, compared to the software processing method, the proposed architecture increases the performance by 21.18 times with 9947 dimensions. The dimension is changeable with a maximum value of 2,097,151, without changing hardware design. This method is also applicable to matrix multiplication processing with other machine learning algorithms.

scholarly journals Values and Consequences in Predictive Machine Evaluation. A Sociology of Predictive Policing

2018 ◽  
Author(s):  
Bilel Benbouzid
Keyword(s):  
Machine Learning ◽  
Research Field ◽  
Machine Learning Algorithms ◽  
Moral Vision ◽  
Predictive Policing ◽  
Method Of Calculation ◽  
Predictive Algorithms ◽  
Cautious Interpretation ◽  
The University ◽  
Two Sides

Predictive policing is a research field whose principal aim is to develop machines for predicting crimes, drawing on machine learning algorithms and the growing availability of a diversity of data. This paper deals with the case of the algorithm of PredPol, the best-known startup in predictive policing. The mathematicians behind it took their inspiration from an algorithm created by a French seismologist, a professor in earth sciences at the University of Savoie. As the source code of the PredPol platform is kept inaccessible as a trade secret, the author contacted the seismologist directly in order to try to understand the predictions of the company’s algorithm. Using the same method of calculation on the same data, the seismologist arrived at a different, more cautious interpretation of the algorithm's capacity to predict crime. How were these predictive analyses formed on the two sides of the Atlantic? How do predictive algorithms come to exist differently in these different contexts? How and why is it that predictive machines can foretell a crime that is yet to be committed in a California laboratory, and yet no longer work in another laboratory in Chambéry? In answering these questions, I found that machine learning researchers have a moral vision of their own activity that can be understood by analyzing the values and material consequences involved in the evaluation tests that are used to create the predictions.

scholarly journals Measurement of Ethical Issues in Software Products

2020 ◽  
Author(s):  
Francesco Di Tria
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Computer Science ◽  
Ethical Issues ◽  
Research Field ◽  
Machine Learning Algorithms ◽  
Software Products ◽  
Ethics Research ◽  
Software Product ◽  
Formal Requirements

Ethics is a research field that is obtaining more and more attention in Computer Science due to the proliferation of artificial intelligence software, machine learning algorithms, robot agents (like chatbot), and so on. Indeed, ethics research has produced till now a set of guidelines, such as ethical codes, to be followed by people involved in Computer Science. However, a little effort has been spent for producing formal requirements to be included in the design process of software able to act ethically with users. In the paper, we investigate those issues that make a software product ethical and propose a set of metrics devoted to quantitatively evaluate if a software product can be considered ethical or not.

scholarly journals Machine Learning based sampling of X-Ray images for a computer-aided detection of Tuberculosis

2021 ◽  
Author(s):  
Fernando Ferreira ◽  
Philipp Gaspar ◽  
Lukas Müller Oliveira ◽  
Rodrigo Torres ◽  
Micael Veríssimo Araújo ◽  
...  
Keyword(s):  
Machine Learning ◽  
Topological Analysis ◽  
Machine Learning Algorithms ◽  
Self Organizing Map ◽  
X Rays ◽  
Computer Aided Detection ◽  
Tuberculosis Screening ◽  
Data Set ◽  
Computer Aided ◽  
Screening Platform

Computer Aided Detection software relies on an annotated data set of X-rays to be developed. The annotation task requires extensive know-how and it is very time-consuming. This work presents a sampling method to select the most relevant images which will be annotated for the development of Tuberculosis screening platform based on machine learning algorithms. The sampling task optimizes the annotation process by reducing the number of images to be analyzed without compromising the diversity and the significance power of the images in the dataset. In this context, the image relevance is based on similarity and dissimilarity measurements. The experiment consisted in a deep learning feature engineering step, followed by topological analysis based on Self-Organizing Map and K-Means.

scholarly journals Accelerating End-to-End Deep Learning for Particle Reconstruction using CMS open data

2021 ◽  
Vol 251 ◽  
pp. 03057
Author(s):  
Michael Andrews ◽  
Bjorn Burkle ◽  
Shravan Chaudhari ◽  
Davide Di Croce ◽  
Sergei Gleyzer ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Top Quark ◽  
Learning Algorithms ◽  
Open Data ◽  
High Energy ◽  
Machine Learning Algorithms ◽  
Multiple Gpus ◽  
Event Identification ◽  
End To End

Machine learning algorithms are gaining ground in high energy physics for applications in particle and event identification, physics analysis, detector reconstruction, simulation and trigger. Currently, most data-analysis tasks at LHC experiments benefit from the use of machine learning. Incorporating these computational tools in the experimental framework presents new challenges. This paper reports on the implementation of the end-to-end deep learning with the CMS software framework and the scaling of the end-to-end deep learning with multiple GPUs. The end-to-end deep learning technique combines deep learning algorithms and low-level detector representation for particle and event identification. We demonstrate the end-to-end implementation on a top quark benchmark and perform studies with various hardware architectures including single and multiple GPUs and Google TPU.

scholarly journals A Smart Architecture for Diabetic Patient Monitoring Using Machine Learning Algorithms

Healthcare ◽  
2020 ◽  
Vol 8 (3) ◽  
pp. 348
Author(s):  
Amine Rghioui ◽  
Jaime Lloret ◽  
Sandra Sendra ◽  
Abdelmajid Oumnad
Keyword(s):  
Machine Learning ◽  
Intelligent System ◽  
Learning Algorithms ◽  
Communication Technologies ◽  
High Energy ◽  
The Body ◽  
Machine Learning Algorithms ◽  
Smart Devices ◽  
Diabetic Patients ◽  
Intelligent Healthcare

Continuous monitoring of diabetic patients improves their quality of life. The use of multiple technologies such as the Internet of Things (IoT), embedded systems, communication technologies, artificial intelligence, and smart devices can reduce the economic costs of the healthcare system. Different communication technologies have made it possible to provide personalized and remote health services. In order to respond to the needs of future intelligent e-health applications, we are called to develop intelligent healthcare systems and expand the number of applications connected to the network. Therefore, the 5G network should support intelligent healthcare applications, to meet some important requirements such as high bandwidth and high energy efficiency. This article presents an intelligent architecture for monitoring diabetic patients by using machine learning algorithms. The architecture elements included smart devices, sensors, and smartphones to collect measurements from the body. The intelligent system collected the data received from the patient, and performed data classification using machine learning in order to make a diagnosis. The proposed prediction system was evaluated by several machine learning algorithms, and the simulation results demonstrated that the sequential minimal optimization (SMO) algorithm gives superior classification accuracy, sensitivity, and precision compared to other algorithms.

Sign in / Sign up

Export Citation Format

Share Document