scholarly journals Using Reinforcement Learning for Generating Polynomial Models to Explain Complex Data

2021 ◽  
Vol 2 (2) ◽  
Author(s):  
Niclas Ståhl ◽  
Gunnar Mathiason ◽  
Dellainey Alcacoas
Keyword(s):  
Machine Learning ◽  
Reinforcement Learning ◽  
Process Parameters ◽  
Machine Learning Algorithms ◽  
Complex Data ◽  
Low Carbon ◽  
Learning To Learn ◽  
Process Data ◽  
Operator Experience ◽  
Prediction Systems

AbstractBasic oxygen steel making is a complex chemical and physical industrial process that reduces a mix of pig iron and recycled scrap into low-carbon steel. Good understanding of the process and the ability to predict how it will evolve requires long operator experience, but this can be augmented with process target prediction systems. Such systems may use machine learning to learn a model of the process based on a long process history, and have an advantage in that they can make use of vastly more process parameters than operators can comprehend. While it has become less of a challenge to build such prediction systems using machine learning algorithms, actual production implementations are rare. The hidden reasoning of complex prediction model and lack of transparency prevents operator trust, even for models that show high accuracy predictions. To express model behaviour and thereby increasing transparency we develop a reinforcement learning (RL) based agent approach, which task is to generate short polynomials that can explain the model of the process from what it has learnt from process data. The RL agent is rewarded on how well it generates polynomials that can predict the process from a smaller subset of the process parameters. Agent training is done with the REINFORCE algorithm, which enables the sampling of multiple concurrently plausible polynomials. Having multiple polynomials, process developers can evaluate several alternative and plausible explanations, as observed in the historic process data. The presented approach gives both a trained generative model and a set of polynomials that can explain the process. The performance of the polynomials is as good as or better than more complex and less interpretable models. Further, the relative simplicity of the resulting polynomials allows good generalisation to fit new instances of data. The best of the resulting polynomials in our evaluation achieves a better $$R^2$$ R 2 score on the test set in comparison to the other machine learning models evaluated.

scholarly journals Heart disease prediction using machine learning techniques : a survey

2018 ◽  
Vol 7 (2.8) ◽  
pp. 684 ◽  
Author(s):  
V V. Ramalingam ◽  
Ayantan Dandapath ◽  
M Karthik Raja
Keyword(s):  
Machine Learning ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Complex Data ◽  
Learning Techniques ◽  
Vector Machines ◽  
Supervised Learning Algorithms ◽  
Life Threatening

Heart related diseases or Cardiovascular Diseases (CVDs) are the main reason for a huge number of death in the world over the last few decades and has emerged as the most life-threatening disease, not only in India but in the whole world. So, there is a need of reliable, accurate and feasible system to diagnose such diseases in time for proper treatment. Machine Learning algorithms and techniques have been applied to various medical datasets to automate the analysis of large and complex data. Many researchers, in recent times, have been using several machine learning techniques to help the health care industry and the professionals in the diagnosis of heart related diseases. This paper presents a survey of various models based on such algorithms and techniques andanalyze their performance. Models based on supervised learning algorithms such as Support Vector Machines (SVM), K-Nearest Neighbour (KNN), NaïveBayes, Decision Trees (DT), Random Forest (RF) and ensemble models are found very popular among the researchers.

Ensemble Methods for APS In-Flight Particle Temperature and Velocity Prediction Considering Torch Electrodes Ageing

2021 ◽  
Author(s):  
K.R. Yu ◽  
C.V. Cojocaru ◽  
F. Ilinca ◽  
E. Irissou
Keyword(s):  
Powder Particle ◽  
Process Parameters ◽  
Particle Temperature ◽  
Ensemble Methods ◽  
Machine Learning Algorithms ◽  
Atmospheric Plasma ◽  
Gradient Boosting ◽  
Input Process ◽  
Process Data ◽  
Particle Characteristics

Abstract In an atmospheric plasma spray (APS) process; in-flight powder particle characteristics; such as the particle velocity and temperature; have significant influence on the coating formation. The nonlinear relationship between the input process parameters and in-flight particle characteristics is thus of paramount importance for coating properties design and quality control. It is also known that the ageing of torch electrodes affects this relationship. In recent years; machine learning algorithms have proven to be able to take into account such complex nonlinear interactions. This work illustrates the application of ensemble methods based on decision tree algorithms to evaluate and to predict in-flight particle temperature and velocity during an APS process considering torch electrodes ageing. Experiments were performed to record simultaneously the input process parameters; the in-flight powder particle characteristics and the electrodes usage time. Various spray durations were considered to emulate industrial coating spray production settings. Random forest and gradient boosting algorithms were used to rank and select the features for the APS process data recorded as the electrodes aged and the corresponding predictive models were compared. The time series aspect of the data will be examined.

scholarly journals Progress in Biomedical Knowledge Discovery: A 25-year Retrospective

2016 ◽  
Vol 25 (S 01) ◽  
pp. S117-S29 ◽  
Author(s):  
L. Sacchi ◽  
J. H. Holmes
Keyword(s):  
Machine Learning ◽  
Knowledge Discovery ◽  
Latent Dirichlet Allocation ◽  
Heterogeneous Data ◽  
Machine Learning Algorithms ◽  
Search Term ◽  
Complex Data ◽  
Learning Approaches ◽  
Biomedical Knowledge ◽  
Observation Period

Summary Objectives: We sought to explore, via a systematic review of the literature, the state of the art of knowledge discovery in biomedical databases as it existed in 1992, and then now, 25 years later, mainly focused on supervised learning. Methods: We performed a rigorous systematic search of PubMed and latent Dirichlet allocation to identify themes in the literature and trends in the science of knowledge discovery in and between time periods and compare these trends. We restricted the result set using a bracket of five years previous, such that the 1992 result set was restricted to articles published between 1987 and 1992, and the 2015 set between 2011 and 2015. This was to reflect the current literature available at the time to researchers and others at the target dates of 1992 and 2015. The search term was framed as: Knowledge Discovery OR Data Mining OR Pattern Discovery OR Pattern Recognition, Automated. Results: A total 538 and 18,172 documents were retrieved for 1992 and 2015, respectively. The number and type of data sources increased dramatically over the observation period, primarily due to the advent of electronic clinical systems. The period 1992-2015 saw the emergence of new areas of research in knowledge discovery, and the refinement and application of machine learning approaches that were nascent or unknown in 1992. Conclusions: Over the 25 years of the observation period, we identified numerous developments that impacted the science of knowledge discovery, including the availability of new forms of data, new machine learning algorithms, and new application domains.Through a bibliometric analysis we examine the striking changes in the availability of highly heterogeneous data resources, the evolution of new algorithmic approaches to knowledge discovery, and we consider from legal, social, and political perspectives possible explanations of the growth of the field. Finally, we reflect on the achievements of the past 25 years to consider what the next 25 years will bring with regard to the availability of even more complex data and to the methods that could be, and are being now developed for the discovery of new knowledge in biomedical data.

scholarly journals Neuromorphic Computation With a Single Magnetic Domain Wall

2021 ◽  
Author(s):  
Razvan V. Ababei ◽  
Matthew O. A. Ellis ◽  
Ian T. Vidamour ◽  
Dhilan S. Devadasan ◽  
Dan A. Allwood ◽  
...  
Keyword(s):  
Magnetic Field ◽  
Machine Learning ◽  
Domain Wall ◽  
Domain Walls ◽  
Magnetic Domain ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Complex Data ◽  
Reservoir Computing ◽  
Model Complex

Abstract Machine learning techniques are commonly used to model complex relationships but implementations on digital hardware are relatively inefficient due to poor matching between conventional computer architectures and the structures of the algorithms they are required to simulate. Neuromorphic devices, and in particular reservoir computing architectures, utilize the inherent properties of physical systems to implement machine learning algorithms and so have the potential to be much more efficient. In this work, we demonstrate that the dynamics of individual domain walls in magnetic nanowires are suitable for implementing the reservoir computing paradigm in hardware. We modelled the dynamics of a domain wall placed between two anti-notches in a nickel nanowire using both a 1d collective coordinates model and micromagnetic simulations. When driven by an oscillating magnetic field, the domain exhibits non-linear dynamics within the potential well created by the anti-notches that are analogous to those of the Duffing oscillator. We exploit the domain wall dynamics for reservoir computing by modulating the amplitude of the applied magnetic field to inject time-multiplexed input signals into the reservoir, and show how this allows us to perform machine learning tasks including: the classification of (1) sine and square waves; (2) spoken digits and (3) non-temporal 2D toy data and hand written digits. Our work lays the foundation for the creation of nanoscale neuromorphic devices in which individual magnetic domain walls are used to perform complex data analysis tasks.

scholarly journals Machine Learning of Optical Properties of Materials - Predicting Spectra from Images and Images from Spectra

2018 ◽  
Author(s):  
Helge S. Stein ◽  
Dan Guevarra ◽  
Paul F Newhouse ◽  
Edwin Soedarmadji ◽  
John Gregoire
Keyword(s):  
Machine Learning ◽  
Optical Properties ◽  
Band Gap ◽  
Materials Science ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Underlying Structure ◽  
Complex Data ◽  
Properties Of Materials ◽  
Optical Properties Of Materials

As the materials science community seeks to capitalize on recent advancements in computer science, the sparsity of well-labelled experimental data and limited throughput by which it can be generated have inhibited deployment of machine learning algorithms to date. Several successful examples in computational chemistry have inspired further adoption of machine learning algorithms, and in the present work we present autoencoding algorithms for measured optical properties of metal oxides, which can serve as an exemplar for the breadth and depth of data required for modern algorithms to learn the underlying structure of experimental materials science data. Our set of 180,902 distinct materials samples spans 78 distinct composition spaces, includes 45 elements, and contains more than 80,000 unique quinary oxide and 67,000 unique quaternary oxide compositions, making it the largest and most diverse experimental materials set utilized in machine learning studies. The extensive dataset enabled training and validation of 3 distinct models for mapping between sample images and absorption spectra, including a conditional variational autoencoder that generates images of hypothetical materials with tailored absorption properties. The absorption patterns auto-generated from sample images capture the salient features of ground truth spectra, and direct band gap energies extracted from these auto-generated patterns are quite accurate with a mean absolute error of 240 meV, which is the approximate uncertainty from traditional extraction of the band gap energy from measurements of the full transmission and reflection spectra. Optical properties of materials are not only ubiquitous in materials applications but also emblematic of the confluence of underlying physical phenomena that yield the type of complex data relationships that merit and benefit from neural network-type modelling.

scholarly journals Application of Machine Learning Algorithms for Solving Problems in the Oil and Gas Sector

2021 ◽  
Vol 19 (3) ◽  
pp. 55-64
Author(s):  
K. N. Maiorov ◽  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Reinforcement Learning ◽  
Machine Learning Algorithms ◽  
Design Stage ◽  
Learning Methods ◽  
Field Development ◽  
Advantages And Disadvantages ◽  
Machine Learning Methods ◽  
The Neural Network

The paper examines the life cycle of field development, analyzes the processes of the field development design stage for the application of machine learning methods. For each process, relevant problems are highlighted, existing solutions based on machine learning methods, ideas and problems are proposed that could be effectively solved by machine learning methods. For the main part of the processes, examples of solutions are briefly described; the advantages and disadvantages of the approaches are identified. The most common solution method is feed-forward neural networks. Subject to preliminary normalization of the input data, this is the most versatile algorithm for regression and classification problems. However, in the problem of selecting wells for hydraulic fracturing, a whole ensemble of machine learning models was used, where, in addition to a neural network, there was a random forest, gradient boosting and linear regression. For the problem of optimizing the placement of a grid of oil wells, the disadvantages of existing solutions based on a neural network and a simple reinforcement learning approach based on Markov decision-making process are identified. A deep reinforcement learning algorithm called Alpha Zero is proposed, which has previously shown significant results in the role of artificial intelligence for games. This algorithm is a decision tree search that directs the neural network: only those branches that have received the best estimates from the neural network are considered more thoroughly. The paper highlights the similarities between the tasks for which Alpha Zero was previously used, and the task of optimizing the placement of a grid of oil producing wells. Conclusions are made about the possibility of using and modifying the algorithm of the optimization problem being solved. Аn approach is proposed to take into account symmetric states in a Monte Carlo tree to reduce the number of required simulations.

scholarly journals Analysis of the possibilities for using machine learning algorithms in the Unity environment

2021 ◽  
Vol 20 ◽  
pp. 197-204
Author(s):  
Karina Litwynenko ◽  
Małgorzata Plechawska-Wójcik
Keyword(s):  
Machine Learning ◽  
Reinforcement Learning ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Learning Performance ◽  
Imitation Learning ◽  
Policy Optimization

Reinforcement learning algorithms are gaining popularity, and their advancement is made possible by the presence of tools to evaluate them. This paper concerns the applicability of machine learning algorithms on the Unity platform using the Unity ML-Agents Toolkit library. The purpose of the study was to compare two algorithms: Proximal Policy Optimization and Soft Actor-Critic. The possibility of improving the learning results by combining these algorithms with Generative Adversarial Imitation Learning was also verified. The results of the study showed that the PPO algorithm can perform better in uncomplicated environments with non-immediate rewards, while the additional use of GAIL can improve learning performance.

scholarly journals Trend analysis of university placement by using machine learning algorithms

2018 ◽  
Vol 7 (2.4) ◽  
pp. 178
Author(s):  
Chandrasekhar Kumbhar ◽  
Dr S. S. Sridhar
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Data Analysis ◽  
Trend Analysis ◽  
Model Building ◽  
Machine Learning Algorithms ◽  
Complex Data ◽  
Student Placement ◽  
The University ◽  
Different Sources

Machine learning is a method of data analysis that automates analytical model building. These models help you to make a trend analysis of university placements data, to predict a placement rate for the students of an upcoming year which will help the university to analyze the performance during placements. Many students look at universities as a means of investment which can help them make a great future by getting placed in good companies and which will relieve their stress and unease from their lives before graduating from the university. The trend will also help in giving the companies reasons as to why they should visit university again and again. Some attributes play the very important role while analyzing the student for e.g. Student’s name, Department, Company, Location and Annual package. So, classification can help you to classify those data and clustering helps to make the clusters department wise. In this paper we have used neural networks to predict the upcoming student placement and got 77% of accuracy while testing were iteration are 1000. Through extensive trend analysis of varies complex data collected from different sources, we can demonstrate that our analysis can provide a good pragmatic solution for future placement of students. 

Machine Learning and Data Mining in Bioinformatics

2012 ◽  
pp. 695-703
Author(s):  
George Tzanis ◽  
Christos Berberidis ◽  
Ioannis Vlahavas
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Data Mining ◽  
Reinforcement Learning ◽  
Supervised Learning ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
The World ◽  
Supervised Learning Algorithms ◽  
Computational Systems

Machine learning is one of the oldest subfields of artificial intelligence and is concerned with the design and development of computational systems that can adapt themselves and learn. The most common machine learning algorithms can be either supervised or unsupervised. Supervised learning algorithms generate a function that maps inputs to desired outputs, based on a set of examples with known output (labeled examples). Unsupervised learning algorithms find patterns and relationships over a given set of inputs (unlabeled examples). Other categories of machine learning are semi-supervised learning, where an algorithm uses both labeled and unlabeled examples, and reinforcement learning, where an algorithm learns a policy of how to act given an observation of the world.

Machine Learning and Data Mining in Bioinformatics

2009 ◽  
pp. 612-621 ◽  
Author(s):  
George Tzanis ◽  
Christos Berberidis ◽  
Ioannis Vlahavas
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Data Mining ◽  
Reinforcement Learning ◽  
Supervised Learning ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
The World ◽  
Supervised Learning Algorithms ◽  
Computational Systems

Machine learning is one of the oldest subfields of artificial intelligence and is concerned with the design and development of computational systems that can adapt themselves and learn. The most common machine learning algorithms can be either supervised or unsupervised. Supervised learning algorithms generate a function that maps inputs to desired outputs, based on a set of examples with known output (labeled examples). Unsupervised learning algorithms find patterns and relationships over a given set of inputs (unlabeled examples). Other categories of machine learning are semi-supervised learning, where an algorithm uses both labeled and unlabeled examples, and reinforcement learning, where an algorithm learns a policy of how to act given an observation of the world.

Sign in / Sign up

Export Citation Format

Share Document