检索结果-内蒙古大学图书馆

Shape-constrained symbolic regression with NSGA-III

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Haider, Christian Josef Ressel Center for Symbolic Regression Heuristic and Evolutionary Algorithms Laboratory University of Applied Sciences Upper Austria Hagenberg Austria

Shape-constrained symbolic regression (SCSR) allows to include prior knowledge into data-based modeling. This inclusion allows to ensure that certain expected behavior is better reflected by the resulting models. The expected behavior is defined via constraints, which refer to the function form e.g. monotonicity, concavity, convexity or the models image boundaries. In addition to the advantage of obtaining more robust and reliable models due to defining constraints over the functions shape, the use of SCSR allows to find models which are more robust to noise and have a better extrapolation behavior. This paper presents a mutlicriterial approach to minimize the approximation error as well as the constraint violations. Explicitly the two algorithms NSGA-II and NSGA-III are implemented and compared against each other in terms of model quality and runtime. Both algorithms are capable of dealing with multiple objectives, whereas NSGA-II is a well established multi-objective approach performing well on instances with up-to 3 objectives. NSGA-III is an extension of the NSGA-II algorithm and was developed to handle problems with"many" objectives (more than 3 objectives). Both algorithms are executed on a selected set of benchmark instances from physics textbooks. The results indicate that both algorithms are able to find largely feasible solutions and NSGA-III provides slight improvements in terms of model quality. Moreover, an improvement in runtime can be observed using the many-objective approach. Copyright © 2022, The Authors. All rights reserved.

关键词： Benchmarking

Concept for a technical infrastructure for management of predictive models in industrial applications

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Bachinger, Florian Kronberge, Gabriel Josef Ressel Center for Symbolic Regression Heuristic and Evolutionary Algorithms Laboratory University of Applied Sciences Upper Austria Hagenberg Austria

With the increasing number of created and deployed prediction models and the complexity of machine learning workflows we require so called model management systems to support data scientists in their tasks. In this work we describe our technological concept for such a model management system. This concept includes versioned storage of data, support for different machine learning algorithms, fine tuning of models, subsequent deployment of models and monitoring of model performance after deployment. We describe this concept with a close focus on model lifecycle requirements stemming from our industry application cases, but generalize key features that are relevant for all applications of machine learning. Copyright © 2021, The Authors. All rights reserved.

关键词： Machine learning

Data aggregation for Reducing training data in symbolic regression

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Kammerer, Lukas Kronberger, Gabriel Kommenda, Michael Josef Ressel Center for Symbolic Regression Heuristic and Evolutionary Algorithms Laboratory University of Applied Sciences Upper Austria Hagenberg Austria

The growing volume of data makes the use of computationally intense machine learning techniques such as symbolic regression with genetic programming more and more impractical. This work discusses methods to reduce the training data and thereby also the runtime of genetic programming. The data is aggregated in a preprocessing step before running the actual machine learning algorithm. K-means clustering and data binning is used for data aggregation and compared with random sampling as the simplest data reduction method. We analyze the achieved speed-up in training and the effects on the trained models' test accuracy for every method on four real-world data sets. The performance of genetic programming is compared with random forests and linear regression. It is shown, that k-means and random sampling lead to very small loss in test accuracy when the data is reduced down to only 30% of the original data, while the speed-up is proportional to the size of the data set. Binning on the contrary, leads to models with very high test error. Copyright © 2021, The Authors. All rights reserved.

关键词： Genetic algorithms

Rank-based Non-dominated Sorting

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Burlacu, Bogdan University of Applied Sciences Upper Austia Austria Heuristic and Evolutionary Algorithms Laboratory Austria Josef Ressel Centre for Symbolic Regression Austria

Non-dominated sorting is a computational bottleneck in Pareto-based multi-objective evolutionary algorithms (MOEAs) due to the runtime-intensive comparison operations involved in establishing dominance relationships between solution candidates. In this paper we introduce Rank Sort, a non-dominated sorting approach exploiting sorting stability and ordinal information to avoid expensive dominance comparisons in the rank assignment phase. Two algorithmic variants are proposed: the first one, RankOrdinal (RO), uses ordinal rank comparisons in order to determine dominance and requires on space;the second one, RankIntersect (RS), uses set intersections and bit-level parallelism and requires on2) space. We demonstrate the efficiency of the proposed methods in comparison with other state of the art algorithms in empirical simulations using the NSGA2 algorithm as well as synthetic benchmarks. The RankIntersect algorithm is able to significantly outperform the current state of the art offering up to 30% speed-up for many objectives. C++ implementations are provided for all algorithms. © 2022, CC BY.

关键词： evolutionary algorithms

Identification of Discrete Non-Linear Dynamics of a Radio-Frequency Power Amplifier Circuit using symbolic regression

学校读者我要写书评

暂无评论

Identification of Discrete Non-Linear Dynamics of a Radio-Fr...

International Symposium on symbolic and Numeric algorithms for Scientific Computing (SYNASC)

作者： Martin Steiger Hans-Georg Brachtendorf Gabriel Kronberger Embedded Systems Lab UAS Upper Austria Hagenberg Austria Josef Ressel Center for Symbolic Regression Heuristic and Evolutionary Algorithms Lab UAS Upper Austria Hagenberg Austria

The identification of non-linearities or undesirable dynamic behavior of electrical components is a common problem. Previous modeling forms are largely based on extensive physical knowledge at the semiconductor level, which has produced reliable solutions over the past decades. This however implies the measurement of physical prototypes in laboratories, which can be costly. It is therefore desirable to have reliable software models of the prototypes available to outsource this procedure to simulators. This paper presents a number of solutions from the field of empirical modeling including symbolic regression, which allow to parameterize such models from measured values. As an example we are utilizing time-domain data from a real radio-frequency power amplifier circuit. We compare a Hammerstein-Wiener model with two methods for symbolic regression, and find that the Hammerstein-Wiener model produces the best predictions but has many non-zero coefficients. Both symbolic regression methods produce short linear models with slightly higher prediction error than the HW model.

关键词： Radio frequency Semiconductor device measurement heuristic algorithms Computational modeling Prototypes Power amplifiers Predictive models

Online Diversity Control in symbolic regression via a Fast Hash-based Tree Similarity Measure

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Burlacu, Bogdan Affenzeller, Michael Kronberger, Gabriel Kommenda, Michael Josef Ressel Centre for Symbolic Regression Heuristic and Evolutionary Algorithms Laboratory University of Applied Sciences Upper Austria Softwarepark 11 Hagenberg4232 Austria

Diversity represents an important aspect of genetic programming, being directly correlated with search performance. When considered at the genotype level, diversity often requires expensive tree distance measures which have a negative impact on the algorithm’s runtime performance. In this work we introduce a fast, hash-based tree distance measure to massively speed-up the calculation of population diversity during the algorithmic run. We combine this measure with the standard GA and the NSGA-II genetic algorithms to steer the search towards higher diversity. We validate the approach on a collection of benchmark problems for symbolic regression where our method consistently outperforms the standard GA as well as NSGA-II configurations with different secondary objectives. Copyright © 2019, The Authors. All rights reserved.

关键词： Genetic programming

Identification of dynamical systems using symbolic regression

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Kronberger, Gabriel Kammerer, Lukas Kommenda, Michael Josef Ressel Centre for Symbolic Regression Heuristic and Evolutionary Algorithms Laboratory University of Applied Sciences Upper Austria Softwarepark 11 Hagenberg4232 Austria

We describe a method for the identification of models for dynamical systems from observational data. The method is based on the concept of symbolic regression and uses genetic programming to evolve a system of ordinary differential equations (ODE). The novelty is that we add a step of gradient-based optimization of the ODE parameters. For this we calculate the sensitivities of the solution to the initial value problem (IVP) using automatic differentiation. The proposed approach is tested on a set of 19 problem instances taken from the literature which includes datasets from simulated systems as well as datasets captured from mechanical systems. We find that gradient-based optimization of parameters improves predictive accuracy of the models. The best results are obtained when we first fit the individual equations to the numeric differences and then subsequently fine-tune the identified parameter values by fitting the IVP solution to the observed variable values. Copyright © 2021, The Authors. All rights reserved.

关键词： Genetic programming

Comparing Shape-Constrained regression algorithms for Data Validation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Bachinger, Florian Kronberger, Gabriel Josef Ressel Center for Symbolic Regression Heuristic and Evolutionary Algorithms Laboratory University of Applied Sciences Upper Austria Hagenberg Austria Johannes Kepler University Linz Austria

Industrial and scientific applications handle large volumes of data that render manual validation by humans infeasible. Therefore, we require automated data validation approaches that are able to consider the prior knowledge of domain experts to produce dependable, trustworthy assessments of data quality. Prior knowledge is often available as rules that describe interactions of inputs with regard to the target e.g. the target must be monotonically decreasing and convex over increasing input values. Domain experts are able to validate multiple such interactions at a glance. However, existing rule-based data validation approaches are unable to consider these constraints. In this work, we compare different shape-constrained regression algorithms for the purpose of data validation based on their classification accuracy and runtime performance. Copyright © 2022, The Authors. All rights reserved.

关键词： regression analysis

Performance of industrial sensor data persistence in data vault 30

学校读者我要写书评

暂无评论

Performance of industrial sensor data persistence in data va...

30th European Modeling and Simulation Symposium, EMSS 2018

作者： Bachinger, Florian Zenisek, Jan Kammerer, Lukas Stimpfl, Martin Kronberger, Gabriel Josef Ressel Centre for Symbolic Regression School of Informatics Communications and Media University of Applied Sciences Upper Austria Hagenberg Austria Heuristic and Evolutionary Algorithms Laboratory University of Applied Sciences Upper Austria Hagenberg Austria Miba AG Dr. Mitterbauer Str. 3 Postfach 3 LaakirchenA-4663 Austria

ISBN: (纸本)9788885741065

Today manufacturing companies are facing important challenges from the market in terms of flexibility, ever growing product mixes, small lot sizes, high competition, etc. To meet these market conditions, digitalization and the use of data are offering a viable toolset considering the advances in the field throughout the last couple of years. The increasing use of sensor technology and the need for interconnecting data from different departments in smart production leads to a surge of recorded data. Persistence and integration of heterogeneous data, generated in a variety of software systems, is a key factor to gain value from data and its analysis. High flexibility in regards to the model is required to accommodate the data. Hence, application of the data vault modelling approach is a fitting candidate to design a data warehouse model. In this paper we present a data vault model for factory sensor data. We analyze the performance of the data warehouse in regards to bulk load of data and common analytic queries. © Institute of Information Science. All rights reserved.

关键词： Data warehouses