Non Linear Fitting Methods for Machine Learning

2017 ◽  
pp. 807-818
Author(s):  
Edgar A. Martínez-García ◽  
Nancy Ávila Rodríguez ◽  
Ricardo Rodríguez-Jorge ◽  
Jolanta Mizera-Pietraszko ◽  
Jaichandar Kulandaidaasan Sheba ◽  
...  
Keyword(s):  
Machine Learning ◽  
Linear Fitting ◽  
Non Linear

scholarly journals A machine-learning-based alloy design platform that enables both forward and inverse predictions for thermo-mechanically controlled processed (TMCP) steel alloys

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Jin-Woong Lee ◽  
Chaewon Park ◽  
Byung Do Lee ◽  
Joonseo Park ◽  
Nam Hoon Goo ◽  
...  
Keyword(s):  
Machine Learning ◽  
Feature Space ◽  
Research Strategy ◽  
Nsga Ii ◽  
Steel Alloys ◽  
Non Linear ◽  
Data Driven Approach ◽  
Processing Information ◽  
Set Up ◽  
World Industry

AbstractPredicting mechanical properties such as yield strength (YS) and ultimate tensile strength (UTS) is an intricate undertaking in practice, notwithstanding a plethora of well-established theoretical and empirical models. A data-driven approach should be a fundamental exercise when making YS/UTS predictions. For this study, we collected 16 descriptors (attributes) that implicate the compositional and processing information and the corresponding YS/UTS values for 5473 thermo-mechanically controlled processed (TMCP) steel alloys. We set up an integrated machine-learning (ML) platform consisting of 16 ML algorithms to predict the YS/UTS based on the descriptors. The integrated ML platform involved regularization-based linear regression algorithms, ensemble ML algorithms, and some non-linear ML algorithms. Despite the dirty nature of most real-world industry data, we obtained acceptable holdout dataset test results such as R2 > 0.6 and MSE < 0.01 for seven non-linear ML algorithms. The seven fully trained non-linear ML models were used for the ensuing ‘inverse design (prediction)’ based on an elitist-reinforced, non-dominated sorting genetic algorithm (NSGA-II). The NSGA-II enabled us to predict solutions that exhibit desirable YS/UTS values for each ML algorithm. In addition, the NSGA-II-driven solutions in the 16-dimensional input feature space were visualized using holographic research strategy (HRS) in order to systematically compare and analyze the inverse-predicted solutions for each ML algorithm.

scholarly journals Physically interpretable machine learning algorithm on multidimensional non-linear fields

2021 ◽  
Vol 428 ◽  
pp. 110074
Author(s):  
Rem-Sophia Mouradi ◽  
Cédric Goeury ◽  
Olivier Thual ◽  
Fabrice Zaoui ◽  
Pablo Tassi
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Machine Learning Algorithm ◽  
Interpretable Machine Learning ◽  
Non Linear

scholarly journals Non-linear fitting with joint spatial regularization in Arterial Spin Labeling

2021 ◽  
pp. 102067
Author(s):  
Oliver Maier ◽  
Stefan M. Spann ◽  
Daniela Pinter ◽  
Thomas Gattringer ◽  
Nicole Hinteregger ◽  
...  
Keyword(s):  
Arterial Spin Labeling ◽  
Spin Labeling ◽  
Linear Fitting ◽  
Non Linear ◽  
Spatial Regularization

scholarly journals Kinetic, isotherm and thermodynamic studies with linear and non-linear fitting for cadmium(II) removal by black carbon of pine cone

2017 ◽  
Vol 76 (8) ◽  
pp. 2242-2253 ◽  
Author(s):  
Z. Yavari ◽  
M. Noroozifar
Keyword(s):  
Aqueous Solutions ◽  
Black Carbon ◽  
Adsorption Process ◽  
Order Kinetic ◽  
Cadmium Removal ◽  
Thermodynamic Studies ◽  
Linear Fitting ◽  
Pine Cone ◽  
Fitting Method ◽  
Non Linear

In this study, black carbon from pine cone (BCPC) and acidic-modified BCPC (MBCPC) powder as a popular agricultural waste in the southeast of Iran were used for cadmium removal from aqueous solutions. The effect of various factors, such as surface chemistry and dosage of adsorbent, contact time, size of particles, initial concentration of cadmium, temperature, and pH of aqueous solutions, was investigated. The results show cadmium removal with usage of the mentioned adsorbents increased after acidic modification. It was noteworthy in this work that the removal percentage of pollutant was above 90% for suggested biosorbents. The obtained experimental data for optimum conditions were selected to model the adsorption behavior of the materials with usage of six isotherm equations via non-linear fitting method and the residual root mean square error estimation for each model. The adsorption of cadmium preferably fitted Khan and Langmuir–Freundlich isotherms for BCPC and MBCPC adsorbents, respectively. The kinetic studies via linear fitting method proved the second-order kinetic was the applicable model for the adsorption process. Thermodynamic studies show the adsorption process of cadmium onto BCPC and MBCPC was spontaneous and endothermic.

A batch-wise non-linear fitting and analysis tool for treating large X-ray diffraction data sets

2006 ◽  
Vol 39 (2) ◽  
pp. 262-266 ◽  
Author(s):  
R. J. Davies
Keyword(s):  
Diffraction Data ◽  
Operation Mode ◽  
Large Data ◽  
Scattering Data ◽  
Data Sets ◽  
Analysis Tool ◽  
Data Set ◽  
X Ray ◽  
Linear Fitting ◽  
Non Linear

Synchrotron sources offer high-brilliance X-ray beams which are ideal for spatially and time-resolved studies. Large amounts of wide- and small-angle X-ray scattering data can now be generated rapidly, for example, during routine scanning experiments. Consequently, the analysis of the large data sets produced has become a complex and pressing issue. Even relatively simple analyses become difficult when a single data set can contain many thousands of individual diffraction patterns. This article reports on a new software application for the automated analysis of scattering intensity profiles. It is capable of batch-processing thousands of individual data files without user intervention. Diffraction data can be fitted using a combination of background functions and non-linear peak functions. To compliment the batch-wise operation mode, the software includes several specialist algorithms to ensure that the results obtained are reliable. These include peak-tracking, artefact removal, function elimination and spread-estimate fitting. Furthermore, as well as non-linear fitting, the software can calculate integrated intensities and selected orientation parameters.

Non-linear fitting-based phasor measurement method for PMU calibrator

2019 ◽  
Vol 106 ◽  
pp. 585-597 ◽  
Author(s):  
Hao Liu ◽  
Sudi Xu ◽  
Tianshu Bi ◽  
Kenneth E. Martin ◽  
Cheng Qian ◽  
...  
Keyword(s):  
Measurement Method ◽  
Linear Fitting ◽  
Phasor Measurement ◽  
Non Linear

Measurement of Cr(VI) Based on Absorption Spectrum through Plastic Fiber

2014 ◽  
Vol 687-691 ◽  
pp. 948-951
Author(s):  
Wei Jun Hu
Keyword(s):  
Signal Processing ◽  
Absorption Spectrum ◽  
Optic Fiber ◽  
Signal Process ◽  
Linear Fitting ◽  
Non Linear ◽  
Low Pass ◽  
Measurement Experiment

Considering the advantage of optic fiber, a methods of measuring Cr (VI) based on absorption spectrum through plastic fiber is introduced, which includes structure of measurement, experiment process, spectrum signal process. After signal processing based on low-pass filtering and non-linear fitting, five concentrations of Cr (VI) can be differed easily and the peak values of spectrum corresponding to five concentrations accord with the Longbow Bill's law . In this way, the measurement concentration can limit down to 0. 0.0660 μg/ml.

Sign in / Sign up

Export Citation Format

Share Document