scholarly journals On the Unfounded Enthusiasm for Soft Selective Sweeps III: The Supervised Machine Learning Algorithm That Isn’t

Genes ◽  
2021 ◽  
Vol 12 (4) ◽  
pp. 527
Author(s):  
Eran Elhaik ◽  
Dan Graur
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
A Priori ◽  
Neutral Theory ◽  
Dominant Mode ◽  
Supervised Machine Learning ◽  
Training Dataset ◽  
Selective Sweeps ◽  
Two Factors ◽  
Negative Controls

In the last 15 years or so, soft selective sweep mechanisms have been catapulted from a curiosity of little evolutionary importance to a ubiquitous mechanism claimed to explain most adaptive evolution and, in some cases, most evolution. This transformation was aided by a series of articles by Daniel Schrider and Andrew Kern. Within this series, a paper entitled “Soft sweeps are the dominant mode of adaptation in the human genome” (Schrider and Kern, Mol. Biol. Evolut. 2017, 34(8), 1863–1877) attracted a great deal of attention, in particular in conjunction with another paper (Kern and Hahn, Mol. Biol. Evolut. 2018, 35(6), 1366–1371), for purporting to discredit the Neutral Theory of Molecular Evolution (Kimura 1968). Here, we address an alleged novelty in Schrider and Kern’s paper, i.e., the claim that their study involved an artificial intelligence technique called supervised machine learning (SML). SML is predicated upon the existence of a training dataset in which the correspondence between the input and output is known empirically to be true. Curiously, Schrider and Kern did not possess a training dataset of genomic segments known a priori to have evolved either neutrally or through soft or hard selective sweeps. Thus, their claim of using SML is thoroughly and utterly misleading. In the absence of legitimate training datasets, Schrider and Kern used: (1) simulations that employ many manipulatable variables and (2) a system of data cherry-picking rivaling the worst excesses in the literature. These two factors, in addition to the lack of negative controls and the irreproducibility of their results due to incomplete methodological detail, lead us to conclude that all evolutionary inferences derived from so-called SML algorithms (e.g., S/HIC) should be taken with a huge shovel of salt.

scholarly journals On the Inapplicability of Supervised Machine Learning to Evolutionary Studies

2020 ◽  
Author(s):  
Eran Elhaik ◽  
Dan Graur
Keyword(s):  
Machine Learning ◽  
A Priori ◽  
Supervised Machine Learning ◽  
Training Dataset ◽  
The Bible ◽  
Human Genomes ◽  
Evolutionary Studies ◽  
Two Factors ◽  
Negative Controls

Supervised machine learning (SML) is a powerful method for predicting a small number of well-defined output groups (e.g., potential buyers of a certain product) by taking as input a large number of known well-defined measurements (e.g., past purchases, income, ethnicity, gender, credit record, age, favorite color, favorite chewing gum). SML is predicated upon the existence of a training dataset in which the correspondence between the input and output is known to be true. SML has had enormous success in the world of commerce, and this success has prompted a few scientists to employ it in the study of molecular and genome evolution. Here, we list the properties of SML that make it an unsuitable tool in evolutionary studies. In particular, we argue that SML cannot be used in an evolutionary exploratory context for the simple reason that training datasets that are known to be a priori true do not exist. As a case study, we use an SML study in which it was concluded that most human genomes evolve by positive selection through soft selective sweeps (Schrider and Kern 2017). We show that in the absence of legitimate training datasets, Schrider and Kern (2017) used (1) simulations that employ many manipulatable variables and (2) a system of cherry-picking data that would put to shame most modern evangelical exegeses of the Bible. These two factors, in addition to the lack of methodological detail and the lack of either negative controls or corrections for multiple comparisons, lead us to conclude that all evolutionary inferences derived from so-called SML algorithms (e.g., discoal) should be taken with a huge shovel of salt.

scholarly journals On the Inapplicability of Supervised Machine Learning to Studying the Driving Forces of Evolution

2021 ◽  
Author(s):  
Eran Elhaik ◽  
Dan Graur
Keyword(s):  
Machine Learning ◽  
Driving Forces ◽  
A Priori ◽  
Supervised Machine Learning ◽  
Training Dataset ◽  
The Bible ◽  
Human Genomes ◽  
Two Factors ◽  
Negative Controls

Supervised machine learning (SML) is a powerful method for predicting a small number of well-defined output groups (e.g., potential buyers of a certain product) by taking as input a large number of known well-defined measurements (e.g., past purchases, income, ethnicity, gender, credit record, age, favorite color, favorite chewing gum). SML is predicated upon the existence of a training dataset in which the correspondence between the input and output is known to be true. SML has had enormous success in the world of commerce, and this success may have prompted a few scientists to employ it in the study of molecular and genome evolution. Here, we list the properties of SML that make it an unsuitable tool in certain evolutionary studies. In particular, we argue that SML cannot be used in an evolutionary exploratory context for the simple reason that training datasets that are known to be a priori true do not exist. As a case study, we use an SML study in which it was concluded that most human genomes evolve by positive selection through soft selective sweeps (Schrider and Kern 2017). We show that in the absence of legitimate training datasets, Schrider and Kern (2017) used (1) simulations that employ many manipulatable variables and (2) a system of cherry-picking data that would put to shame most modern evangelical exegeses of the Bible. These two factors, in addition to the lack of methodological detail and negative controls, lead us to conclude that all evolutionary inferences derived from so-called SML algorithms (e.g., S-HIC) should be taken with a huge shovel of salt.

Using Supervised Machine Learning Algorithms for Automated Lithology Prediction from Wireline Log Data

2021 ◽  
Author(s):  
Marian Popescu ◽  
Rebecca Head ◽  
Tim Ferriday ◽  
Kate Evans ◽  
Jose Montero ◽  
...  
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Machine Learning Algorithms ◽  
Supervised Machine Learning ◽  
Training Dataset ◽  
Depth Interval ◽  
Log Data ◽  
Machine Learning Approach ◽  
Lithology Prediction ◽  
Logging While Drilling

Abstract This paper presents advancements in machine learning and cloud deployment that enable rapid and accurate automated lithology interpretation. A supervised machine learning technique is described that enables rapid, consistent, and accurate lithology prediction alongside quantitative uncertainty from large wireline or logging-while-drilling (LWD) datasets. To leverage supervised machine learning, a team of geoscientists and petrophysicists made detailed lithology interpretations of wells to generate a comprehensive training dataset. Lithology interpretations were based on applying determinist cross-plotting by utilizing and combining various raw logs. This training dataset was used to develop a model and test a machine learning pipeline. The pipeline was applied to a dataset previously unseen by the algorithm, to predict lithology. A quality checking process was performed by a petrophysicist to validate new predictions delivered by the pipeline against human interpretations. Confidence in the interpretations was assessed in two ways. The prior probability was calculated, a measure of confidence in the input data being recognized by the model. Posterior probability was calculated, which quantifies the likelihood that a specified depth interval comprises a given lithology. The supervised machine learning algorithm ensured that the wells were interpreted consistently by removing interpreter biases and inconsistencies. The scalability of cloud computing enabled a large log dataset to be interpreted rapidly; >100 wells were interpreted consistently in five minutes, yielding >70% lithological match to the human petrophysical interpretation. Supervised machine learning methods have strong potential for classifying lithology from log data because: 1) they can automatically define complex, non-parametric, multi-variate relationships across several input logs; and 2) they allow classifications to be quantified confidently. Furthermore, this approach captured the knowledge and nuances of an interpreter's decisions by training the algorithm using human-interpreted labels. In the hydrocarbon industry, the quantity of generated data is predicted to increase by >300% between 2018 and 2023 (IDC, Worldwide Global DataSphere Forecast, 2019–2023). Additionally, the industry holds vast legacy data. This supervised machine learning approach can unlock the potential of some of these datasets by providing consistent lithology interpretations rapidly, allowing resources to be used more effectively.

scholarly journals Detecting Plasma Detachment in the Wendelstein 7-X Stellarator Using Machine Learning

2021 ◽  
Vol 12 (1) ◽  
pp. 269
Author(s):  
Máté Szűcs ◽  
Tamás Szepesi ◽  
Christoph Biedermann ◽  
Gábor Cseh ◽  
Marcin Jakubowski ◽  
...  
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Supervised Machine Learning ◽  
Training Dataset ◽  
Computational Time ◽  
Learning Models ◽  
Camera System ◽  
Pixel Intensity ◽  
On The Road ◽  
Machine Learning Models

The detachment regime has a high potential to play an important role in fusion devices on the road to a fusion power plant. Complete power detachment has been observed several times during the experimental campaigns of the Wendelstein 7-X (W7-X) stellarator. Automatic observation and signaling of such events could help scientists to better understand these phenomena. With the growing discharge times in fusion devices, machine learning models and algorithms are a powerful tool to process the increasing amount of data. We investigate several classical supervised machine learning models to detect complete power detachment in the images captured by the Event Detection Intelligent Camera System (EDICAM) at the W7-X at each given image frame. In the dedicated detached state the plasma is stable despite its reduced contact with the machine walls and the radiation belt stays close to the separatrix, without exhibiting significant heat load onto the divertor. To decrease computational time and resources needed we propose certain pixel intensity profiles (or intensity values along lines) as the input to these models. After finding the profile that describes the images best in terms of detachment, we choose the best performing machine learning algorithm. It achieves an F1 score of 0.9836 on the training dataset and 0.9335 on the test set. Furthermore, we investigate its predictions in other scenarios, such as plasmas with substantially decreased minor radius and several magnetic configurations.

scholarly journals Predictive Modelling of Employee Turnover in Indian IT Industry Using Machine Learning Techniques

2019 ◽  
Vol 23 (1) ◽  
pp. 12-21 ◽  
Author(s):  
Shikha N. Khera ◽  
Divya
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Confusion Matrix ◽  
Predictive Modelling ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
It Industry ◽  
Knowledge Based ◽  
Employee Attrition

Information technology (IT) industry in India has been facing a systemic issue of high attrition in the past few years, resulting in monetary and knowledge-based loses to the companies. The aim of this research is to develop a model to predict employee attrition and provide the organizations opportunities to address any issue and improve retention. Predictive model was developed based on supervised machine learning algorithm, support vector machine (SVM). Archival employee data (consisting of 22 input features) were collected from Human Resource databases of three IT companies in India, including their employment status (response variable) at the time of collection. Accuracy results from the confusion matrix for the SVM model showed that the model has an accuracy of 85 per cent. Also, results show that the model performs better in predicting who will leave the firm as compared to predicting who will not leave the company.

scholarly journals Deep Learning Classification of Canine Behavior Using a Single Collar-Mounted Accelerometer: Real-World Validation

Animals ◽  
2021 ◽  
Vol 11 (6) ◽  
pp. 1549
Author(s):  
Robert D. Chambers ◽  
Nathanael C. Yoder ◽  
Aletha B. Carson ◽  
Christian Junge ◽  
David E. Allen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Real World ◽  
Learning Algorithm ◽  
Drinking Behavior ◽  
True Positive Rate ◽  
Training Dataset ◽  
Activity Levels ◽  
Accelerometer Data ◽  
Activity Monitors

Collar-mounted canine activity monitors can use accelerometer data to estimate dog activity levels, step counts, and distance traveled. With recent advances in machine learning and embedded computing, much more nuanced and accurate behavior classification has become possible, giving these affordable consumer devices the potential to improve the efficiency and effectiveness of pet healthcare. Here, we describe a novel deep learning algorithm that classifies dog behavior at sub-second resolution using commercial pet activity monitors. We built machine learning training databases from more than 5000 videos of more than 2500 dogs and ran the algorithms in production on more than 11 million days of device data. We then surveyed project participants representing 10,550 dogs, which provided 163,110 event responses to validate real-world detection of eating and drinking behavior. The resultant algorithm displayed a sensitivity and specificity for detecting drinking behavior (0.949 and 0.999, respectively) and eating behavior (0.988, 0.983). We also demonstrated detection of licking (0.772, 0.990), petting (0.305, 0.991), rubbing (0.729, 0.996), scratching (0.870, 0.997), and sniffing (0.610, 0.968). We show that the devices’ position on the collar had no measurable impact on performance. In production, users reported a true positive rate of 95.3% for eating (among 1514 users), and of 94.9% for drinking (among 1491 users). The study demonstrates the accurate detection of important health-related canine behaviors using a collar-mounted accelerometer. We trained and validated our algorithms on a large and realistic training dataset, and we assessed and confirmed accuracy in production via user validation.

scholarly journals Study on Influence of Range of Data in Concrete Compressive Strength with Respect to the Accuracy of Machine Learning with Linear Regression

2021 ◽  
Vol 11 (9) ◽  
pp. 3866
Author(s):  
Jun-Ryeol Park ◽  
Hye-Jin Lee ◽  
Keun-Hyeok Yang ◽  
Jung-Keun Kook ◽  
Sanghee Kim
Keyword(s):  
Machine Learning ◽  
Compressive Strength ◽  
Linear Regression ◽  
Coarse Aggregate ◽  
Learning Algorithm ◽  
Linear Regression Analysis ◽  
Aggregate Size ◽  
Training Dataset ◽  
Machine Learning Algorithm ◽  
Concrete Compressive Strength

This study aims to predict the compressive strength of concrete using a machine-learning algorithm with linear regression analysis and to evaluate its accuracy. The open-source software library TensorFlow was used to develop the machine-learning algorithm. In the machine-earning algorithm, a total of seven variables were set: water, cement, fly ash, blast furnace slag, sand, coarse aggregate, and coarse aggregate size. A total of 4297 concrete mixtures with measured compressive strengths were employed to train and testing the machine-learning algorithm. Of these, 70% were used for training, and 30% were utilized for verification. For verification, the research was conducted by classifying the mixtures into three cases: the case where the machine-learning algorithm was trained using all the data (Case-1), the case where the machine-learning algorithm was trained while maintaining the same number of training dataset for each strength range (Case-2), and the case where the machine-learning algorithm was trained after making the subcase of each strength range (Case-3). The results indicated that the error percentages of Case-1 and Case-2 did not differ significantly. The error percentage of Case-3 was far smaller than those of Case-1 and Case-2. Therefore, it was concluded that the range of training dataset of the concrete compressive strength is as important as the amount of training dataset for accurately predicting the concrete compressive strength using the machine-learning algorithm.

scholarly journals Identification of abnormal tribological regimes using a microphone and semi-supervised machine-learning algorithm

Friction ◽  
2021 ◽  
Author(s):  
Vigneashwara Pandiyan ◽  
Josef Prost ◽  
Georg Vorlaufer ◽  
Markus Varga ◽  
Kilian Wasmer
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Industrial Applications ◽  
Supervised Machine Learning ◽  
Normal Operation ◽  
Promising Alternative ◽  
Wear And Tear ◽  
Validation Procedure ◽  
And Performance

AbstractFunctional surfaces in relative contact and motion are prone to wear and tear, resulting in loss of efficiency and performance of the workpieces/machines. Wear occurs in the form of adhesion, abrasion, scuffing, galling, and scoring between contacts. However, the rate of the wear phenomenon depends primarily on the physical properties and the surrounding environment. Monitoring the integrity of surfaces by offline inspections leads to significant wasted machine time. A potential alternate option to offline inspection currently practiced in industries is the analysis of sensors signatures capable of capturing the wear state and correlating it with the wear phenomenon, followed by in situ classification using a state-of-the-art machine learning (ML) algorithm. Though this technique is better than offline inspection, it possesses inherent disadvantages for training the ML models. Ideally, supervised training of ML models requires the datasets considered for the classification to be of equal weightage to avoid biasing. The collection of such a dataset is very cumbersome and expensive in practice, as in real industrial applications, the malfunction period is minimal compared to normal operation. Furthermore, classification models would not classify new wear phenomena from the normal regime if they are unfamiliar. As a promising alternative, in this work, we propose a methodology able to differentiate the abnormal regimes, i.e., wear phenomenon regimes, from the normal regime. This is carried out by familiarizing the ML algorithms only with the distribution of the acoustic emission (AE) signals captured using a microphone related to the normal regime. As a result, the ML algorithms would be able to detect whether some overlaps exist with the learnt distributions when a new, unseen signal arrives. To achieve this goal, a generative convolutional neural network (CNN) architecture based on variational auto encoder (VAE) is built and trained. During the validation procedure of the proposed CNN architectures, we were capable of identifying acoustics signals corresponding to the normal and abnormal wear regime with an accuracy of 97% and 80%. Hence, our approach shows very promising results for in situ and real-time condition monitoring or even wear prediction in tribological applications.

Sign in / Sign up

Export Citation Format

Share Document