scholarly journals Fast mode decomposition in few-mode fibers

2020 ◽  
Vol 11 (1) ◽  
Author(s):  
Egor S. Manuylovich ◽  
Vladislav V. Dvoyrin ◽  
Sergei K. Turitsyn
Keyword(s):  
Machine Learning ◽  
Intensity Distribution ◽  
High Performance ◽  
Phase Retrieval ◽  
Low Cost ◽  
Industrial Applications ◽  
Machine Learning Techniques ◽  
Decomposition Algorithms ◽  
Measured Intensity ◽  
Mode Decomposition

Abstract Retrieval of the optical phase information from measurement of intensity is of a high interest because this would facilitate simple and cost-efficient techniques and devices. In scientific and industrial applications that exploit multi-mode fibers, a prior knowledge of spatial mode structure of the fiber, in principle, makes it possible to recover phases using measured intensity distribution. However, current mode decomposition algorithms based on the analysis of the intensity distribution at the output of a few-mode fiber, such as optimization methods or neural networks, still have high computational costs and high latency that is a serious impediment for applications, such as telecommunications. Speed of signal processing is one of the key challenges in this approach. We present a high-performance mode decomposition algorithm with a processing time of tens of microseconds. The proposed mathematical algorithm that does not use any machine learning techniques, is several orders of magnitude faster than the state-of-the-art deep-learning-based methods. We anticipate that our results can stimulate further research on algorithms beyond popular machine learning methods and they can lead to the development of low-cost phase retrieval receivers for various applications of few-mode fibers ranging from imaging to telecommunications.

scholarly journals Systematic Methods on Machine Learning Techniques for Clinical Predictive Modelling

2020 ◽  
Vol 9 (5) ◽  
pp. 288-297
Keyword(s):  
Machine Learning ◽  
Data Mining ◽  
High Performance ◽  
Predictive Modelling ◽  
Medical Application ◽  
Medical Data ◽  
Machine Learning Techniques ◽  
Screening Process ◽  
Huge Data ◽  
System Data

Predictive modelling is a mathematical technique which uses Statistics for prediction, due to the rapid growth of data over the cloud system, data mining plays a significant role. Here, the term data mining is a way of extracting knowledge from huge data sources where it’s increasing the attention in the field of medical application. Specifically, to analyse and extract the knowledge from both known and unknown patterns for effective medical diagnosis, treatment, management, prognosis, monitoring and screening process. But the historical medical data might include noisy, missing, inconsistent, imbalanced and high dimensional data.. This kind of data inconvenience lead to severe bias in predictive modelling and decreased the data mining approach performances. The various pre-processing and machine learning methods and models such as Supervised Learning, Unsupervised Learning and Reinforcement Learning in recent literature has been proposed. Hence the present research focuses on review and analyses the various model, algorithm and machine learning technique for clinical predictive modelling to obtain high performance results from numerous medical data which relates to the patients of multiple diseases.

scholarly journals Supermarket Sales Prediction Using Regression

2021 ◽  
Vol 10 (2) ◽  
pp. 1153-1157
Keyword(s):  
Machine Learning ◽  
Performance Metrics ◽  
Low Cost ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Customer Data ◽  
Sales Data ◽  
Online Marketplace ◽  
Sales Prediction ◽  
The Future

Sales forecasting is an important when it comes to companies who are engaged in retailing, logistics, manufacturing, marketing and wholesaling. It allows companies to allocate resources efficiently, to estimate revenue of the sales and to plan strategies which are better for company’s future. In this paper, predicting product sales from a particular store is done in a way that produces better performance compared to any machine learning algorithms. The dataset used for this project is Big Mart Sales data of the 2013.Nowadays shopping malls and Supermarkets keep track of the sales data of the each and every individual item for predicting the future demand of the customer. It contains large amount of customer data and the item attributes. Further, the frequent patterns are detected by mining the data from the data warehouse. Then the data can be used for predicting the sales of the future with the help of several machine learning techniques (algorithms) for the companies like Big Mart. In this project, we propose a model using the Xgboost algorithm for predicting sales of companies like Big Mart and founded that it produces better performance compared to other existing models. An analysis of this model with other models in terms of their performance metrics is made in this project. Big Mart is an online marketplace where people can buy or sell or advertise your merchandise at low cost. The goal of the paper is to make Big Mart the shopping paradise for the buyers and a marketing solutions for the sellers as well. The ultimate aim is the complete satisfaction of the customers. The project “SUPERMARKET SALES PREDICTION” builds a predictive model and finds out the sales of each of the product at a particular store. The Big Mart use this model to under the properties of the products which plays a major role in increasing the sales. This can also be done on the basis hypothesis that should be done before looking at the data

scholarly journals Implementation of machine learning techniques to control the speed of the vehicles

2018 ◽  
Vol 7 (4.5) ◽  
pp. 654
Author(s):  
M. S. Satyanarayana ◽  
Aruna T.M ◽  
Divyaraj G.N
Keyword(s):  
Machine Learning ◽  
Low Cost ◽  
Machine Learning Techniques ◽  
New Technique ◽  
Vehicle Speed ◽  
Learning Techniques ◽  
Traffic Awareness ◽  
Alarm Sound ◽  
The Government ◽  
A New Technique

Accidents have become major issue in Developing countries like India now a day. As per the Surveys 60% of the accidents are happening due to over speed. Though the government has taken so many initiatives like Traffic Awareness & Driving Awareness Week etc.., but still the percentage of accidents are not getting reduced. In this paper a new technique has been introduced to reduce the percentage of accidents. The new technique is implemented using the concept of Machine Learning [1]. The Machine Learning based systems can be implemented in all vehicles to avoid the accidents at low cost [1]. The main objective of this system is to calculate the speed of the vehicle at three various locations based on the place where the vehicle speed must be controlled and if the speed is greater than the designated speed in that road then the vehicle automatically detects the problem and same will be intimated to the driver to control the speed of the vehicle. If the speed is less or equal to the designated speed in that road then the vehicle will be passed without any disturbance. The system will be giving beep sound along with color indication to driver in each and every scenario. The other option implemented in this system is if the driver is driving the vehicle in the night and if he feel drowsy the system detects it immediately and alarm sound will be initiated to wake up the driver. This system though it won’t avoid 100% accidents at least it will reduce the percentage of accidents. This system is not only to avoid accidents it will also intelligently control the speed of the vehicles and creates awareness amongst the drivers.  

scholarly journals Environmental Sound Recognition on Embedded Systems: From FPGAs to TPUs

Electronics ◽  
2021 ◽  
Vol 10 (21) ◽  
pp. 2622
Author(s):  
Jurgen Vandendriessche ◽  
Nick Wouters ◽  
Bruno da Silva ◽  
Mimoun Lamrini ◽  
Mohamed Yassin Chkouri ◽  
...  
Keyword(s):  
Machine Learning ◽  
High Performance ◽  
Machine Learning Techniques ◽  
Sound Recognition ◽  
Learning Approaches ◽  
Environmental Sound ◽  
Embedded Devices ◽  
Power Efficient ◽  
Computationally Intensive ◽  
Environmental Sound Recognition

In recent years, Environmental Sound Recognition (ESR) has become a relevant capability for urban monitoring applications. The techniques for automated sound recognition often rely on machine learning approaches, which have increased in complexity in order to achieve higher accuracy. Nonetheless, such machine learning techniques often have to be deployed on resource and power-constrained embedded devices, which has become a challenge with the adoption of deep learning approaches based on Convolutional Neural Networks (CNNs). Field-Programmable Gate Arrays (FPGAs) are power efficient and highly suitable for computationally intensive algorithms like CNNs. By fully exploiting their parallel nature, they have the potential to accelerate the inference time as compared to other embedded devices. Similarly, dedicated architectures to accelerate Artificial Intelligence (AI) such as Tensor Processing Units (TPUs) promise to deliver high accuracy while achieving high performance. In this work, we evaluate existing tool flows to deploy CNN models on FPGAs as well as on TPU platforms. We propose and adjust several CNN-based sound classifiers to be embedded on such hardware accelerators. The results demonstrate the maturity of the existing tools and how FPGAs can be exploited to outperform TPUs.

Using AI at the Edge and Incremental Machine Learning to Process Onboard Instrument Data

2021 ◽  
Author(s):  
Nicholas Parkyn
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Real Time ◽  
Data Storage ◽  
High Performance ◽  
Heterogeneous Computing ◽  
Low Cost ◽  
Machine Intelligence ◽  
Edge Computing ◽  
Continual Learning

Emerging heterogeneous computing, computing at the edge, machine learning and AI at the edge technology drives approaches and techniques for processing and analysing onboard instrument data in near real-time. The author has used edge computing and neural networks combined with high performance heterogeneous computing platforms to accelerate AI workloads. Heterogeneous computing hardware used is readily available, low cost, delivers impressive AI performance and can run multiple neural networks in parallel. Collecting, processing and machine learning from onboard instruments data in near real-time is not a trivial problem due to data volumes, complexities of data filtering, data storage and continual learning. Little research has been done on continual machine learning which aims at a higher level of machine intelligence through providing the artificial agents with the ability to learn from a non-stationary and never-ending stream of data. The author has applied the concept of continual learning to building a system that continually learns from actual boat performance and refines predictions previously done using static VPP data. The neural networks used are initially trained using the output from traditional VPP software and continue to learn from actual data collected under real sailing conditions. The author will present the system design, AI, and edge computing techniques used and the approaches he has researched for incremental training to realise continual learning.

A reverse design model for high-performance and low-cost magnesium alloys by machine learning

2022 ◽  
Vol 201 ◽  
pp. 110881
Author(s):  
Xiaoxi Mi ◽  
Lianjuan Tian ◽  
Aitao Tang ◽  
Jing Kang ◽  
Peng Peng ◽  
...  
Keyword(s):  
Machine Learning ◽  
Magnesium Alloys ◽  
High Performance ◽  
Low Cost ◽  
Design Model

Explore the Use of Handwriting Information and Machine Learning Techniques in Evaluating Mental Workload

2020 ◽  
pp. 1423-1439
Author(s):  
Zhiming Wu ◽  
Tao Lin ◽  
Ningjiu Tang
Keyword(s):  
Machine Learning ◽  
Interaction Design ◽  
Mental Stress ◽  
Mental Workload ◽  
Low Cost ◽  
Research Question ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Learning Techniques ◽  
Psychological Evidence

Mental workload is considered one of the most important factors in interaction design and how to detect a user's mental workload during tasks is still an open research question. Psychological evidence has already attributed a certain amount of variability and “drift” in an individual's handwriting pattern to mental stress, but this phenomenon has not been explored adequately. The intention of this paper is to explore the possibility of evaluating mental workload with handwriting information by machine learning techniques. Machine learning techniques such as decision trees, support vector machine (SVM), and artificial neural network were used to predict mental workload levels in the authors' research. Results showed that it was possible to make prediction of mental workload levels automatically based on handwriting patterns with relatively high accuracy, especially on patterns of children. In addition, the proposed approach is attractive because it requires no additional hardware, is unobtrusive, is adaptable to individual users, and is of very low cost.

Drought Stress Detection Using Low-Cost Computer Vision Systems and Machine Learning Techniques

2020 ◽  
Vol 22 (3) ◽  
pp. 27-29 ◽  
Author(s):  
Paula Ramos-Giraldo ◽  
Chris Reberg-Horton ◽  
Anna M. Locke ◽  
Steven Mirsky ◽  
Edgar Lobaton
Keyword(s):  
Machine Learning ◽  
Computer Vision ◽  
Drought Stress ◽  
Low Cost ◽  
Machine Learning Techniques ◽  
Stress Detection ◽  
Vision Systems ◽  
Learning Techniques ◽  
Computer Vision Systems

A one-pot route to the synthesis of alloyed Cu/Ag bimetallic nanoparticles with different mass ratios for catalytic reduction of 4-nitrophenol

2015 ◽  
Vol 3 (7) ◽  
pp. 3450-3455 ◽  
Author(s):  
Wei Wu ◽  
Mei Lei ◽  
Shuanglei Yang ◽  
Li Zhou ◽  
Li Liu ◽  
...  
Keyword(s):  
Bimetallic Catalysts ◽  
High Performance ◽  
Bimetallic Nanoparticles ◽  
Catalytic Reduction ◽  
Low Cost ◽  
Industrial Applications ◽  
Research Interest ◽  
Alloy Nanoparticles ◽  
One Pot ◽  
Copper Based Alloy

Copper-based alloy nanoparticles (NPs) have recently triggered much research interest for the development of low-cost and high-performance bimetallic catalysts that have industrial applications.

Sign in / Sign up

Export Citation Format

Share Document