Multicore Embedded Worst-Case Task Design Issues and Analysis Using Machine Learning Logic

The field of biosciences have advanced to a larger extent and have generated large amounts of information from Electronic Health Records. This have given rise to the acute need of knowledge generation from this enormous amount of data. Data mining methods and machine learning play a major role in this aspect of biosciences. Chronic Kidney Disease(CKD) is a condition in which the kidneys are damaged and cannot filter blood as they always do. A family history of kidney diseases or failure, high blood pressure, type 2 diabetes may lead to CKD. This is a lasting damage to the kidney and chances of getting worser by time is high. The very common complications that results due to a kidney failure are heart diseases, anemia, bone diseases, high potasium and calcium. The worst case situation leads to complete kidney failure and necessitates kidney transplant to live. An early detection of CKD can improve the quality of life to a greater extent. This calls for good prediction algorithm to predict CKD at an earlier stage . Literature shows a wide range of machine learning algorithms employed for the prediction of CKD. This paper uses data preprocessing,data transformation and various classifiers to predict CKD and also proposes best Prediction framework for CKD. The results of the framework show promising results of better prediction at an early stage of CKD

Download Full-text

A trajectories' guide to the state space - learning missing terms in bifurcating ecological systems

10.5194/egusphere-egu21-16159 ◽

2021 ◽

Author(s):

Rahel Vortmeyer-Kley ◽

Pascal Nieters ◽

Gordon Pipa

Keyword(s):

Machine Learning ◽

Differential Equations ◽

Algal Blooms ◽

Ecological Systems ◽

Short Term ◽

Specific System ◽

Worst Case ◽

System Components ◽

Full Knowledge

Ecological systems typically can exhibit various states ranging from extinction to coexistence of different species in oscillatory states. The switch from one state to another is called bifurcation. All these behaviours of a specific system are hidden in a set of describing differential equations (DE) depending on different parametrisations. To model such a system as DE requires full knowledge of all possible interactions of the system components. In practise, modellers can end up with terms in the DE that do not fully describe the interactions or in the worst case with missing terms.The framework of universal differential equations (UDE) for scientific machine learning (SciML) [1] allows to reconstruct the incomplete or missing term from an idea of the DE and a short term timeseries of the system and make long term predictions of the system&#8217;s behaviour. However, the approach in [1] has difficulties to reconstruct the incomplete or missing term in systems with bifurcations. We developed a trajectory-based loss metric for UDE and SciML to tackle the problem and tested it successfully on a system mimicking algal blooms in the ocean.[1] Rackauckas, Christopher, et al. "Universal differential equations for scientific machine learning." arXiv preprint arXiv:2001.04385 (2020).

Download Full-text

Identification of NLOS and Multi-Path Conditions in UWB Localization Using Machine Learning Methods

Applied Sciences ◽

10.3390/app10113980 ◽

2020 ◽

Vol 10 (11) ◽

pp. 3980 ◽

Cited By ~ 4

Author(s):

Cung Lian Sang ◽

Bastian Steinhagen ◽

Jonas Dominik Homburg ◽

Michael Adams ◽

Marc Hesse ◽

...

Keyword(s):

Machine Learning ◽

Indoor Localization ◽

Measurement Data ◽

Ultra Wideband ◽

Machine Learning Techniques ◽

Line Of Sight ◽

Support Vector ◽

Case Scenario ◽

Worst Case ◽

Worst Case Scenario

In ultra-wideband (UWB)-based wireless ranging or distance measurement, differentiation between line-of-sight (LOS), non-line-of-sight (NLOS), and multi-path (MP) conditions is important for precise indoor localization. This is because the accuracy of the reported measured distance in UWB ranging systems is directly affected by the measurement conditions (LOS, NLOS, or MP). However, the major contributions in the literature only address the binary classification between LOS and NLOS in UWB ranging systems. The MP condition is usually ignored. In fact, the MP condition also has a significant impact on the ranging errors of the UWB compared to the direct LOS measurement results. However, the magnitudes of the error contained in MP conditions are generally lower than completely blocked NLOS scenarios. This paper addresses machine learning techniques for identification of the three mentioned classes (LOS, NLOS, and MP) in the UWB indoor localization system using an experimental dataset. The dataset was collected in different conditions in different scenarios in indoor environments. Using the collected real measurement data, we compared three machine learning (ML) classifiers, i.e., support vector machine (SVM), random forest (RF) based on an ensemble learning method, and multilayer perceptron (MLP) based on a deep artificial neural network, in terms of their performance. The results showed that applying ML methods in UWB ranging systems was effective in the identification of the above-three mentioned classes. Specifically, the overall accuracy reached up to 91.9% in the best-case scenario and 72.9% in the worst-case scenario. Regarding the F1-score, it was 0.92 in the best-case and 0.69 in the worst-case scenario. For reproducible results and further exploration, we provide the publicly accessible experimental research data discussed in this paper at PUB (Publications at Bielefeld University). The evaluations of the three classifiers are conducted using the open-source Python machine learning library scikit-learn.

Download Full-text

Machine-Learning-Enabled DDoS Attacks Detection in P4 Programmable Networks

Journal of Network and Systems Management ◽

10.1007/s10922-021-09633-5 ◽

2021 ◽

Vol 30 (1) ◽

Author(s):

Francesco Musumeci ◽

Ali Can Fidanci ◽

Francesco Paolucci ◽

Filippo Cugini ◽

Massimo Tornatore

Keyword(s):

Machine Learning ◽

Real Time ◽

Transmission Control Protocol ◽

Denial Of Service ◽

Attack Detection ◽

Ddos Attacks ◽

Worst Case ◽

Latency Reduction ◽

Data Plane ◽

Network Switches

Abstract Distributed Denial of Service (DDoS) attacks represent a major concern in modern Software Defined Networking (SDN), as SDN controllers are sensitive points of failures in the whole SDN architecture. Recently, research on DDoS attacks detection in SDN has focused on investigation of how to leverage data plane programmability, enabled by P4 language, to detect attacks directly in network switches, with marginal involvement of SDN controllers. In order to effectively address cybersecurity management in SDN architectures, we investigate the potential of Artificial Intelligence and Machine Learning (ML) algorithms to perform automated DDoS Attacks Detection (DAD), specifically focusing on Transmission Control Protocol SYN flood attacks. We compare two different DAD architectures, called Standalone and Correlated DAD, where traffic features collection and attack detection are performed locally at network switches or in a single entity (e.g., in SDN controller), respectively. We combine the capability of ML and P4-enabled data planes to implement real-time DAD. Illustrative numerical results show that, for all tested ML algorithms, accuracy, precision, recall and F1-score are above 98% in most cases, and classification time is in the order of few hundreds of $$\upmu \text {s}$$ μ s in the worst case. Considering real-time DAD implementation, significant latency reduction is obtained when features are extracted at the data plane by using P4 language. Graphic Abstract

Download Full-text

Cached Sufficient Statistics for Efficient Machine Learning with Large Datasets

Journal of Artificial Intelligence Research ◽

10.1613/jair.453 ◽

1998 ◽

Vol 8 ◽

pp. 67-91 ◽

Cited By ~ 93

Author(s):

A. Moore ◽

M. S. Lee

Keyword(s):

Machine Learning ◽

Data Structures ◽

Rule Learning ◽

Worst Case ◽

Sufficient Statistics ◽

Frequent Sets ◽

Efficient Machine ◽

Real World Datasets ◽

Selection Algorithms ◽

New Algorithms

This paper introduces new algorithms and data structures for quick counting for machine learning datasets. We focus on the counting task of constructing contingency tables, but our approach is also applicable to counting the number of records in a dataset that match conjunctive queries. Subject to certain assumptions, the costs of these operations can be shown to be independent of the number of records in the dataset and loglinear in the number of non-zero entries in the contingency table. We provide a very sparse data structure, the ADtree, to minimize memory use. We provide analytical worst-case bounds for this structure for several models of data distribution. We empirically demonstrate that tractably-sized data structures can be produced for large real-world datasets by (a) using a sparse tree structure that never allocates memory for counts of zero, (b) never allocating memory for counts that can be deduced from other counts, and (c) not bothering to expand the tree fully near its leaves. We show how the ADtree can be used to accelerate Bayes net structure finding algorithms, rule learning algorithms, and feature selection algorithms, and we provide a number of empirical results comparing ADtree methods against traditional direct counting approaches. We also discuss the possible uses of ADtrees in other machine learning methods, and discuss the merits of ADtrees in comparison with alternative representations such as kd-trees, R-trees and Frequent Sets.

Download Full-text

Further investigation of 3D dose verification in proton therapy utilizing acoustic signal, wavelet decomposition and machine learning

Biomedical Physics & Engineering Express ◽

10.1088/2057-1976/ac396d ◽

2021 ◽

Author(s):

Songhuan Yao ◽

Zongsheng Hu ◽

Qiang Xie ◽

Yidong Yang ◽

Hao Peng

Keyword(s):

Machine Learning ◽

Proton Therapy ◽

Acoustic Signal ◽

Wavelet Decomposition ◽

Short Term Memory ◽

Three Dimensions ◽

Dose Verification ◽

Case Scenario ◽

Worst Case ◽

Dose Distributions

Abstract Online dose verification in proton therapy is a critical task for quality assurance. We further studied the feasibility of using a wavelet-based machine learning framework to accomplishing that goal in three dimensions, built upon our previous work in 1D. The wavelet decomposition was utilized to extract features of acoustic signals and a bidirectional long-short-term memory (Bi-LSTM) recurrent neural network (RNN) was used. The 3D dose distributions of mono-energetic proton beams (multiple beam energies) inside a 3D CT phantom, were generated using Monte-Carlo simulation. The 3D propagation of acoustic signal was modeled using the k-Wave toolbox. Three different beamlets (i.e. acoustic pathways) were tested, one with its own model. The performance was quantitatively evaluated in terms of mean relative error (MRE) of dose distribution and positioning error of Bragg peak (△BP ), for two signal-to-noise ratios (SNRs). Due to the lack of experimental data for the time being, two SNR conditions were modeled (SNR=1 and 5). The model is found to yield good accuracy and noise immunity for all three beamlets. The results exhibit an MRE below 0.6% (without noise) and 1.2% (SNR= 5), and △BP below 1.2 mm (without noise) and 1.3 mm (SNR= 5). For the worst-case scenario (SNR=1), the MRE and △BP are below 2.3% and 1.9 mm, respectively. It is encouraging to find out that our model is able to identify the correlation between acoustic waveforms and dose distributions in 3D heterogeneous tissues, as in the 1D case. The work lays a good foundation for us to advance the study and fully validate the feasibility with experimental results.

Download Full-text

Motion Inference Using Sparse Inertial Sensors, Self-Supervised Learning, and a New Dataset of Unscripted Human Motion

Sensors ◽

10.3390/s20216330 ◽

2020 ◽

Vol 20 (21) ◽

pp. 6330

Author(s):

Jack H. Geissinger ◽

Alan T. Asbeck

Keyword(s):

Machine Learning ◽

Inertial Sensors ◽

Wearable Sensors ◽

Human Motion ◽

Supervised Machine Learning ◽

Upper Body ◽

Worst Case ◽

Open World ◽

Fitness Training ◽

Full Body

In recent years, wearable sensors have become common, with possible applications in biomechanical monitoring, sports and fitness training, rehabilitation, assistive devices, or human-computer interaction. Our goal was to achieve accurate kinematics estimates using a small number of sensors. To accomplish this, we introduced a new dataset (the Virginia Tech Natural Motion Dataset) of full-body human motion capture using XSens MVN Link that contains more than 40 h of unscripted daily life motion in the open world. Using this dataset, we conducted self-supervised machine learning to do kinematics inference: we predicted the complete kinematics of the upper body or full body using a reduced set of sensors (3 or 4 for the upper body, 5 or 6 for the full body). We used several sequence-to-sequence (Seq2Seq) and Transformer models for motion inference. We compared the results using four different machine learning models and four different configurations of sensor placements. Our models produced mean angular errors of 10–15 degrees for both the upper body and full body, as well as worst-case errors of less than 30 degrees. The dataset and our machine learning code are freely available.

Download Full-text

A Novel Fast Searching Algorithm Based on Least Square Regression

Revue d intelligence artificielle ◽

10.18280/ria.350111 ◽

2021 ◽

Vol 35 (1) ◽

pp. 93-98

Author(s):

Ratna Kumari Challa ◽

Siva Prasad Chintha ◽

B. Reddaiah ◽

Kanusu Srinivasa Rao

Keyword(s):

Machine Learning ◽

Predictive Modelling ◽

Binary Search ◽

Least Square ◽

Search Performance ◽

Worst Case ◽

Data Set ◽

Least Square Regression ◽

Interpolation Search ◽

Fast Searching

Currently, the machine learning group is well-understood and commonly used for predictive modelling and feature generation through linear methodologies such as reversals, principal analysis and canonical correlation analyses. All these approaches are typically intended to capture fascinating subspaces in the original space of high dimensions. These methods have all a closed-form approach because of its simple linear structures, which makes estimation and theoretical analysis for small datasets very straightforward. However, it is very common for a data set to have millions or trillions of samples and features in modern machine learning problems. We deal with the problem of fast estimation from large volumes of data for ordinary squares. The search operation is a very important operation and it is useful in many applications. Some applications when the data set size is large, the linear search takes the time which is proportional to the size of the data set. Binary search and interpolation search performs good for the search of elements in the data set in O(logn) and ⋅O(log(⋅logn)) respectively in the worst case. Now, in this paper, an effort is made to develop a novel fast searching algorithm based on the least square regression curve fitting method. The algorithm is implemented and its execution results are analyzed and compared with binary search and interpolation search performance. The proposed model is compared with the traditional methods and the proposed fast searching algorithm exhibits better performance than the traditional models.

Download Full-text

Coupling Between the Modeling and Controller-Design Problems—Part I: Analysis

Journal of Dynamic Systems Measurement and Control ◽

10.1115/1.2801285 ◽

1997 ◽

Vol 119 (3) ◽

pp. 498-502 ◽

Cited By ~ 11

Author(s):

G. A. Brusher ◽

P. T. Kabamba ◽

A. G. Ulsoy

Keyword(s):

General Framework ◽

Controller Design ◽

Companion Paper ◽

Design Approach ◽

Design Issues ◽

Worst Case ◽

Design Problems ◽

Intuitive Notion ◽

Performance Specifications

Controller-design issues influence modeling decisions; therefore, the modeling and controller-design problems are coupled. Part I of this paper proposes a quantification of this coupling in terms of the size of the set of models from which satisfactory controllers may be derived. Moreover, in the context of an extremely general framework, we prove that the coupling increases as performance specifications are tightened and as uncertainty is increased. Furthermore, the existence of a preferred model for controller design is demonstrated; this model is preferred in the sense that its use leads to improved robustness or performance. However, this model does not necessarily satisfy our intuitive notion of a worst-case model; rather, it depends on the performance specifications and uncertainty. The analysis of coupling presented herein motivates the design approach formulated in the companion paper, Part II.

Download Full-text

Augmenting Pre-Analysis Plans with Machine Learning

AEA Papers and Proceedings ◽

10.1257/pandp.20191070 ◽

2019 ◽

Vol 109 ◽

pp. 71-76 ◽

Cited By ~ 1

Author(s):

Jens Ludwig ◽

Sendhil Mullainathan ◽

Jann Spiess

Keyword(s):

Machine Learning ◽

Worst Case ◽

Ex Post ◽

Spurious Results

Concerns about the dissemination of spurious results have led to calls for pre-analysis plans (PAPs) to avoid ex-post “p-hacking.” But often the conceptual hypotheses being tested do not imply the level of specificity required for a PAP. In this paper we suggest a framework for PAPs that capitalize on the availability of causal machine-learning (ML) techniques, in which researchers combine specific aspects of the analysis with ML for the flexible estimation of unspecific remainders. A “cheap-lunch” result shows that the inclusion of ML produces limited worst-case costs in power, while offering a substantial upside from systematic specification searches.

Download Full-text