Learning Algorithm For Estimation Research Articles

Benchmarking machine learning algorithms for adaptive quantum phase estimation with noisy intermediate-scale quantum sensors

Quantum phase estimation is a paradigmatic problem in quantum sensing and metrology. Here we show that adaptive methods based on classical machine learning algorithms can be used to enhance the precision of quantum phase estimation when noisy non-entangled qubits are used as sensors. We employ the Differential Evolution (DE) and Particle Swarm Optimization (PSO) algorithms to this task and we identify the optimal feedback policies which minimize the Holevo variance. We benchmark these schemes with respect to scenarios that include Gaussian and Random Telegraph fluctuations as well as reduced Ramsey-fringe visibility due to decoherence. We discuss their robustness against noise in connection with real experimental setups such as Mach–Zehnder interferometry with optical photons and Ramsey interferometry in trapped ions, superconducting qubits and nitrogen-vacancy (NV) centers in diamond.

Review of PPG signal using Machine Learning Algorithms for Blood Pressure and Glucose Estimation

This paper presents an innovative technique for current non-invasive measurement of systolic(SBP) and diastolic (DBP) blood pressures and BGL (blood glucose level). Blood glucose and blood pressure are the most significant factors which marks the health issues, adequate measurement of these parameters are requires by a vast range of people. This paper focuses that the measurement of these parameters can effectively and accurately achieved through photoplethysmography(PPG) which is the one among the non-invasive methods. The analysis of the PPG signals are also made to the check of the accuracy of the device.This review paper focuses on understanding the BP-related features from PPG and explores the growth of this technology in terms of validation, sample size, diversity of topics, based on the datasets used over the period between 2010 - 2019. The data are preprocessed through the normal machine learning techniques and the algorithm of artificial intelligence and neural networks are applied into it. From this analysis, the accuracy of the data is also checked into. All these methods are used for the continuous monitoring and evaluation of the blood pressure and glucose level using PPG signals in a non-invasive ways.

Sub-2 mm depth of interaction localization in PET detectors with prismatoid light guide arrays and single-ended readout using convolutional neural networks.

Depth of interaction (DOI) readout in PET imaging has been researched in efforts to mitigate parallax error, which would enable the development of small diameter, high-resolution PET scanners. However, DOI PET has not yet been commercialized due to the lack of practical, cost-effective, and data efficient DOI readout methods. The rationale for this study was to develop a supervised machine learning algorithm for DOI estimation in PET that can be trained and deployed on unique sets of crystals. Depth collimated flood data was experimentally acquired using a Na-22 source with a depth-encoding single-ended readout Prism-PET module consisting of lutetium yttrium orthosilicate (LYSO) crystals coupled 4-to-1 to 3×3 silicon photomultiplier (SiPM) pixels on one end and a segmented prismatoid light guide array on the other end. A convolutional neural network (CNN) was trained to perform DOI estimation on data from center, edge and corner crystals in the Prism-PET module using (a) all non-zero readout pixels and (b) only the 4 highest readout signals per event. CNN testing was performed on data from crystals not included in CNN training. An average DOI resolution of 1.84mm full width at half maximum (FWHM) across all crystals was achieved when using all readout signals per event with the CNN compared to 3.04mm FWHM DOI resolution using classical estimation. When using only the 4 highest signals per event, an average DOI resolution of 1.92mm FWHM was achieved, representing only a 4% dropoff in CNN performance compared to using all non-zero pixels per event. Our CNN-based DOI estimation algorithm provides the best reported DOI resolution in a single-ended readout module and can be readily deployed on crystals not used for model training.

Bayesian Iterative Channel Estimation and Turbo Equalization for Multiple-Input–Multiple-Output Underwater Acoustic Communications

This article investigates a robust receiver scheme for a single carrier, multiple-input-multiple-output (MIMO) underwater acoustic (UWA) communications, which uses the sparse Bayesian learning algorithm for iterative channel estimation embedded in Turbo equalization (TEQ). We derive a block-wise sparse Bayesian learning framework modeling the spatial correlation of the MIMO UWA channels, where a more robust expectation-maximization algorithm is proposed for updating the joint estimates of channel impulse response, residual noise, and channel covariance matrix. By exploiting the spatially correlated sparsity of MIMO UWA channels and the second-order a priori channel statistics from the training sequence, the proposed Bayesian channel estimator enjoys not only relatively low complexity but also more stable control of the hyperparameters that determine the channel sparsity and recovery accuracy. Moreover, this article proposes a low complexity space-time soft decision feedback equalizer (ST-SDFE) with successive soft interference cancellation. Evaluated by the undersea 2008 Surface Processes and Acoustic Communications Experiment, the improved sparse Bayesian learning channel estimation algorithm outperforms the conventional Bayesian algorithms in terms of the robustness and complexity, while enjoying better estimation accuracy than the orthogonal matching pursuit and the improved proportionate normalized least mean squares algorithms. We have also verified that the proposed ST-SDFE TEQ significantly outperforms the low-complexity minimum mean square error TEQ in terms of the bit error rate and error propagation.

Recurrent Neural Networks and Acoustic Features for Frame-Level Signal-to-Noise Ratio Estimation

It is important to know the presence and the relative level of background noise for many speech processing tasks. Frame-level signal-to-noise ratio (SNR) provides a measure of instantaneous noise level of a noisy signal, and its estimation has been researched for decades. This problem can be approached from a supervised learning perspective by predicting SNR from features of noisy speech. In this study, we introduce a deep learning algorithm for frame-level SNR estimation. The proposed algorithm employs recurrent neural networks (RNNs) with long short-term memory (LSTM) to leverage contextual information. We also systematically examine a range of acoustic features and investigate feature combinations using Group Lasso and sequential floating forward selection (SFFS). The proposed algorithm naturally leads to an utterance-level SNR estimator. Systematical evaluations show that the proposed algorithm provides an accurate estimate of frame-level SNR, as well as utterance-level SNR, under different noise conditions, outperforming other estimators.

A machine learning-based framework for predicting game server load

Server load prediction can be utilized for load-balancing and load-sharing in distributed systems. The use of machine learning (ML) algorithms for load estimation in distributed system applications can increase the availability and performance of servers. Hence, a number of machine learning algorithms have been applied thus far for server load estimation. This study focuses on increasing the performance of game servers by accurately predicting the workload of game servers in short, medium and long term prediction situations. While doing this, various machine learning techniques have been applied and the algorithms that give the best results are presented. In terms of implementation, companies using their servers and data centers can try to increase their level of satisfaction by using these algorithms. A prediction model is developed and the estimation performances of a number of fundamental ML methods i.e., Naive Bayes (NB), Generalized Linear Model (GLM), Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Gradient Boosted Trees (GBT), Support Vector Machine (SVM), Fast Large Margin (FLM), Convolutional Neural Network CNN are analyzed. The data used during the training stage is obtained by listening to the TCP/IP packet traffic and the real-data is extracted by performing an extensive analysis of the total transferred-data that includes also the payload. In the analysis phase, the goodput is considered in order to reveal exact resource requirements. Comprehensive simulations are performed under various conditions for high accuracy performance analysis. Experimental results indicate that the proposed ML-based prediction shows promising performance in terms of load prediction when compared to the common approaches present in the literature.

A Review on Drone-Based Data Solutions for Cereal Crops

Food security is a longstanding global issue over the last few centuries. Eradicating hunger and all forms of malnutrition by 2030 is still a key challenge. The COVID-19 pandemic has placed additional stress on food production, demand, and supply chain systems; majorly impacting cereal crop producer and importer countries. Short food supply chain based on the production from local farms is less susceptible to travel and export bans and works as a smooth system in the face of these stresses. Local drone-based data solutions can provide an opportunity to address these challenges. This review aims to present a deeper understanding of how the drone-based data solutions can help to combat food insecurity caused due to the pandemic, zoonotic diseases, and other food shocks by enhancing cereal crop productivity of small-scale farming systems in low-income countries. More specifically, the review covers sensing capabilities, promising algorithms, and methods, and added-value of novel machine learning algorithms for local-scale monitoring, biomass and yield estimation, and mapping of them. Finally, we present the opportunities for linking information from citizen science, internet of things (IoT) based on low-cost sensors and drone-based information to satellite data for upscaling crop yield estimation to a larger geographical extent within the Earth Observation umbrella.

A Model-Driven Deep Learning Algorithm for Joint Activity Detection and Channel Estimation

This letter provides a deep learning framework for massive grant-free random access in 6G cellular internet of things (IoT) networks. A model-driven deep learning algorithm for joint activity detection and channel estimation is proposed based on the principle of approximate massage passing (AMP). This algorithm only needs to learn four key parameters, but not the whole algorithm architecture. More importantly, it does not require the prior information about active probabilities and channel variance, and can significantly improve the performance with a finite number of training data. Simulation results validate the effectiveness of the proposed deep learning algorithm.

A novel adaptive deep learning model of Covid-19 with focus on mortality reduction strategies.

We employ deep learning to propose an Artificial Neural Network (ANN) based and data stream guided real-time incremental learning algorithm for parameter estimation of a non-intrusive, intelligent, adaptive and online analytical model of Covid-19 disease. Modeling and simulation of such problems pose an additional challenge of continuously evolving training data in which the model parameters change over time depending upon external factors. Our main contribution is that in a scenario of continuously evolving training data, unlike typical deep learning techniques, this non-intrusive algorithm eliminates the need to retrain or rebuild the model from scratch every time a new training data set is received. After validating the model, we use it to study the impact of different strategies for epidemic control. Finally, we propose and simulate a strategy of controlled natural immunization through risk-based population compartmentalization (PC) wherein the population is divided in Low Risk (LR) and High Risk (HR) compartments based on risk factors (like comorbidities and age) and subjected to different disease transmission dynamics by isolating the HR compartment while allowing the LR compartment to develop natural immunity. Upon release from the preventive isolation, the HR compartment finds itself surrounded by enough number of immunized individuals to prevent the spread of infection and thus most of the deaths occurring in this group are avoided.

Predictive analytics approaches for software effort estimation: A review

Background/Objective: In Software Effort Estimation (SEE), predicting the amount of time taken in human hours or months for software development is considered as a cumbersome process. SEE consists of both Software Development Effort Estimation (SDEE) and Software Maintenance Effort Estimation (SMEE). Over estimation or under estimation of software effort results in project cancellation or project failure. The objective of this study is to identify the best performing model for software Effort Estimation through experimental comparison with various Machine learning algorithms. Methods: Software Effort Estimation was addressed by using various machine learning techniques such as Multilinear Regression, Ridge Regression, Lasso Regression, ElasticNet Regression, Random Forest, Support Vector Machine, Decision Tree and NeuralNet to recognize best performing model. Datasets used are Desharnais, Maxwell, China and Albrecht datasets. Evaluation metrics considered are Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Square Error (RMSE) and R-Squared. Findings: Experiments on various machine learning algorithms for software Effort Estimation determines that Support Vector Machine produced the best performance comparatively with other algorithms. Keywords: Machine learning; software effort estimation; regression models; classification models

Computer vision system for measuring individual cow feed intake using RGB-D camera and deep learning algorithms

This study entailed the design and implementation of a computer vision system for cow individual feed intake measurement, based on deep Convolutional Neural Networks (CNNs) models, and a low-cost RGB-D (Red, Green, Blue, Depth) camera. Individual feed intake of dairy cows is an important variable currently unavailable in commercial dairies. An RGB-D camera was positioned above the feeding area in an open cowshed. Feed intake was estimated by combining information from the RGB and depth images. Cow identification was conducted using the RGB image. Deep learning algorithms for identification and intake estimation were developed using CNN models. Data for CNN training were acquired by a specially developed automatic data acquisition system. A range of feed weights under varied configurations were collected over a period of seven days with the setup, which included an automatic scale, cameras, and a micro-controller. Test data for feed intake was acquired in an open cowshed research dairy farm, wherein the cows were fed Total Mix Ration (TMR). Images of cows eating over a period of 36 h provided the test data for cow identification. The system was able to accurately identify 93.65% of the cows. The amount of feed consumed, which ranged from 0 to 8 kg per meal, was measured with mean absolute and square errors (MAE and MSE) of 0.127 kg, and 0.034kg2 respectively. The analysis showed that the amount and diversity of data are important for model training. Better results were achieved for the model that was trained with high-diversity data than the model trained with homogeneous data (MAE of 1.025 kg, and MSE of 2.845 kg2 for a model trained on shadow conditions only). Additionally, the training analysis shows that the model based on RGB-D data shows better results than the model based on depth channel data without RGB (MAE of 0.241 kg, and MSE of 0.106 kg2). These results suggest the potential of low-cost cameras for individual feed intake measurements in advanced dairy farms.

Interference Mitigation for Coexisting Wireless Body Area Networks: Distributed Learning Solutions

When multiple wireless body area networks (WBANs) exist in close proximity to each other, the inter-user interference considerably degrades the signal to interference plus noise ratio of the packets arriving at each WBAN coordinator. Also, the propagation paths within each WBAN experience fading due to the continuous changes in the body posture and mobility of the human body. The most preferred coexisting mechanisms specified in the IEEE 802.15.6 standard is the channel hopping mechanism, which fails to consider the varying radio environment and obtained reward in its channel selection. Thus, our paper investigates this channel selection problem for interference mitigation in a time-varying environment. We formulate this channel selection problem as a finite repeated potential game and propose two learning algorithms, Stochastic Learning Algorithm (SLA) and Stochastic Estimator Learning Algorithm (SELA) to achieve the Nash Equilibrium (NE) of the game. Numerical results show the convergence of the learning algorithms to the NE point of the game. The performance evaluation and impact of parameters on these two algorithms are also analyzed in our paper.

Prediction of Safe-Zone Evolution in Poisonous Environment Based on Deep Learning

Precise online identification of safe-zone evolution has been long desired in the context of indoor chemical attack events. Computational fluid dynamics (CFD) technology can provide great accuracy but is incapable of applying on online predictions of a large-scale fluid system due to the enormous computational costs to date. In this paper, we propose a Multi-Step Spatial-Temporal Situational-Awareness Network (MSSTP-SA Net) based on deep learning algorithms for rapid online estimation of concentration field. The dataset is firstly created by CFD simulation considering different poisonous gas release and decontamination scenarios in pre-specific domain with various combinations of airflow conditions and source parameters. Corresponding experiments are also provided for validations. The test experiments of the trained network suggest that the evolution of concentration field's distribution can be predicted faithfully in millisecond computation time costs. We hope this approach to be highly useful in most chemical attack scenarios to reduce casualties.

P4347Automated echocardiographic quantification of left ventricular ejection fraction without volume measurements using a machine learning algorithm mimicking a human expert

Abstract Background Echocardiographic quantification of left ventricular (LV) ejection fraction (EF) relies on either manual or automated identification of endocardial boundaries followed by standard calculation of model-based end-systolic and end-diastolic LV volumes. Recent developments in artificial intelligence resulted in computer algorithms that allow near automated detection of endocardial boundaries and measurement of LV volumes and function. However, boundary identification is still prone to errors limiting accuracy in certain patients. We hypothesized that a fully automated machine learning algorithm could be developed, which circumvents border detection and instead estimates the degree of ventricular contraction, similar to a human expert trained on tens of thousands of images. Purpose This study was designed to test the feasibility and accuracy of this approach. Methods Machine learning algorithm was developed and trained on a database of >50,000 echocardiographic studies, including multiple apical 2- and 4-chamber views, to automatically estimate LVEF (AutoEF, BayLabs). Testing was performed on an independent group of 99 unselected patients, whose automated EF values were compared to reference values obtained by averaging measurements by 3 experts using conventional volume-based technique. Inter-technique agreement was assessed using linear regression and Bland-Altman analysis of bias and limits of agreement (LOA). Consistency was assessed by mean absolute deviation (MAD) among automated estimates based on different combinations of apical views. Finally, sensitivity and specificity of detecting of EF≤35% was calculated. These metrics were compared side-by-side against the same reference standard to those obtained from conventional EF measurements by clinical readers. Results Automated estimation of LVEF was feasible in all 99 patients. AutoEF values showed high consistency (MAD=2.9%) and excellent agreement with the reference values: r=0.95, bias=1.0%, LOA=±11.8%, with sensitivity 0.90 and specificity 0.92 for detection of EF≤35%. This was similar to clinicians' measurements: r=0.94, bias=1.4%, LOA=±13.4%,sensitivity 0.93, specificity 0.87. Conclusions Machine learning algorithm for volume-independent LVEF estimation is highly feasible and similar in accuracy to conventional volume-based measurements, when compared to reference values provided by an expert panel. Acknowledgement/Funding Bay Labs, Inc.

Combined state and parameter estimation in level-set methods

Reduced-order models based on level-set methods are widely used tools to qualitatively capture and track the nonlinear dynamics of an interface. The aim of this paper is to develop a physics-informed, data-driven, statistically rigorous learning algorithm for state and parameter estimation with level-set methods. A Bayesian approach based on data assimilation is introduced. Data assimilation is enabled by the ensemble Kalman filter and smoother, which are used in their probabilistic formulations. The level-set data assimilation framework is verified in one-dimensional and two-dimensional test cases, where state estimation, parameter estimation and uncertainty quantification are performed. The statistical performance of the proposed ensemble Kalman filter and smoother is quantified by twin experiments. In the twin experiments, the combined state and parameter estimation fully recovers the reference solution, which validates the proposed algorithm. The level-set data assimilation framework is then applied to the prediction of the nonlinear dynamics of a forced premixed flame, which exhibits the formation of sharp cusps and intricate topological changes, such as pinch-off events. The proposed physics-informed statistical learning algorithm opens up new possibilities for making reduced-order models of interfaces quantitatively predictive, any time that reference data is available.

Automated Echocardiographic Quantification of Left Ventricular Ejection Fraction Without Volume Measurements Using a Machine Learning Algorithm Mimicking a Human Expert.

Echocardiographic quantification of left ventricular (LV) ejection fraction (EF) relies on either manual or automated identification of endocardial boundaries followed by model-based calculation of end-systolic and end-diastolic LV volumes. Recent developments in artificial intelligence resulted in computer algorithms that allow near automated detection of endocardial boundaries and measurement of LV volumes and function. However, boundary identification is still prone to errors limiting accuracy in certain patients. We hypothesized that a fully automated machine learning algorithm could circumvent border detection and instead would estimate the degree of ventricular contraction, similar to a human expert trained on tens of thousands of images. Machine learning algorithm was developed and trained to automatically estimate LVEF on a database of >50 000 echocardiographic studies, including multiple apical 2- and 4-chamber views (AutoEF, BayLabs). Testing was performed on an independent group of 99 patients, whose automated EF values were compared with reference values obtained by averaging measurements by 3 experts using conventional volume-based technique. Inter-technique agreement was assessed using linear regression and Bland-Altman analysis. Consistency was assessed by mean absolute deviation among automated estimates from different combinations of apical views. Finally, sensitivity and specificity of detecting of EF ≤35% were calculated. These metrics were compared side-by-side against the same reference standard to those obtained from conventional EF measurements by clinical readers. Automated estimation of LVEF was feasible in all 99 patients. AutoEF values showed high consistency (mean absolute deviation =2.9%) and excellent agreement with the reference values: r=0.95, bias=1.0%, limits of agreement =±11.8%, with sensitivity 0.90 and specificity 0.92 for detection of EF ≤35%. This was similar to clinicians' measurements: r=0.94, bias=1.4%, limits of agreement =±13.4%, sensitivity 0.93, specificity 0.87. Machine learning algorithm for volume-independent LVEF estimation is highly feasible and similar in accuracy to conventional volume-based measurements, when compared with reference values provided by an expert panel.

Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories.

Rodent skilled reaching is commonly used to study dexterous skills, but requires significant time and effort to implement the task and analyze the behavior. Several automated versions of skilled reaching have been developed recently. Here, we describe a version that automatically presents pellets to rats while recording high-definition video from multiple angles at high frame rates (300 fps). The paw and individual digits are tracked with DeepLabCut, a machine learning algorithm for markerless pose estimation. This system can also be synchronized with physiological recordings, or be used to trigger physiologic interventions (e.g., electrical or optical stimulation).

CFAR Block-Sparse Bayesian Learning Algorithm for the Off-grid DOA Estimation with Coprime Array

Constant false-alarm rate Block-sparse Bayesian learning (CFAR-BSBL) algorithm is proposed to reduce the computational complexity and improve Direction of arrival (DOA) estimation accuracy of offgrid signals with coprime array. Firstly, a signal model with normalized noise is built to avoid the learning procedure of noise parameter. Secondly, a block sparse Bayesian framework is built with the introduction of a temporary correlation matrix in order to use t he temporal structure of incident signals. Then the algorithm uses CFAR detection to detect the grids close to the real DOA and relieve the dependence on the number of signals. Finally, an off-grid process based on the closest grids is adopted to deal with the off-grid problem. The proposed CFAR-BSBL algorithm can obtain high accuracy and low complexity DOA estimation of off-grid signals with coprime array.

Application of feature selection methods and machine learning algorithms for saltmarsh biomass estimation using Worldview-2 imagery

Assessing large scale plant productivity of coastal marshes is essential to understand the resilience of these systems to climate change. Two machine learning approaches, random forest (RF) and support vector machine (SVM) regression were tested to estimate biomass of a common saltmarshes species, salt couch grass (Sporobolus virginicus). Reflectance and vegetation indices derived from 8 bands of Worldview-2 multispectral data were used for four experiments to develop the biomass model. These four experiments were, Experiment-1: 8 bands of Worldview-2 image, Experiment-2: Possible combination of all bands of Worldview-2 for Normalized Difference Vegetation Index (NDVI) type vegetation indices, Experiment-3: Combination of bands and vegetation indices, Experiment-4: Selected variables derived from experiment-3 using variable selection methods. The main objectives of this study are (i) to recommend an affordable low cost data source to predict biomass of a common saltmarshes species, (ii) to suggest a variable selection method suitable for multispectral data, (iii) to assess the performance of RF and SVM for the biomass prediction model. Cross-validation of parameter optimizations for SVM showed that optimized parameter of ɛ-SVR failed to provide a reliable prediction. Hence, ν-SVR was used for the SVM model. Among the different variable selection methods, recursive feature elimination (RFE) selected a minimum number of variables (only 4) with an RMSE of 0.211 (kg/m2). Experiment-4 (only selected bands) provided the best results for both of the machine learning regression methods, RF (R 2= 0.72, RMSE= 0.166 kg/m2) and SVR (R 2= 0.66, RMSE = 0.200 kg/m2) to predict biomass. When a 10-fold cross validation of the RF model was compared with a 10-fold cross validation of SVR, a significant difference (p = <0.0001) was observed for RMSE. One to one comparisons of actual to predicted biomass showed that RF underestimates the high biomass values, whereas SVR overestimates the values; this suggests a need for further investigation and refinement.

Deep Learning-Based Channel Estimation

In this letter, we present a deep learning algorithm for channel estimation in communication systems. We consider the time–frequency response of a fast fading communication channel as a 2D image. The aim is to find the unknown values of the channel response using some known values at the pilot locations. To this end, a general pipeline using deep image processing techniques, image super-resolution (SR), and image restoration (IR) is proposed. This scheme considers the pilot values, altogether, as a low-resolution image and uses an SR network cascaded with a denoising IR network to estimate the channel. Moreover, the implementation of the proposed pipeline is presented. The estimation error shows that the presented algorithm is comparable to the minimum mean square error (MMSE) with full knowledge of the channel statistics, and it is better than an approximation to linear MMSE. The results confirm that this pipeline can be used efficiently in channel estimation.