Investigating the parameter space of evolutionary algorithms

Sipper, Moshe; Fu, Weixuan; Ahuja, Karuna; Moore, Jason H.

doi:10.1186/s13040-018-0164-x

Cited by 68 publications

(39 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These values have been shown to produce accurate results, and since a preliminary tuning phase (performed with a grid search) showed no (statistically significant) change in terms of performance with respect to other configurations tested, we relied on these values. This is related to the fact that it has become increasingly clear that GP is very robust to parameter values once a good configuration is determined, as suggested by a recent and in-depth study [33]. Again, as anticipated in Section 3, these results confirm that all the considered classifiers achieve a very similar performance on unseen data, even if the training performance is better for some of them.…”

Section: Resultssupporting

confidence: 76%

A Scalable Genetic Programming Approach to Integrate miRNA‐Target Predictions: Comparing Different Parallel Implementations of M3GP

et al. 2018

View full text Add to dashboard Cite

There are many molecular biology approaches to the analysis of microRNA (miRNA) and target interactions, but the experiments are complex and expensive. For this reason, in silico computational approaches able to model these molecular interactions are highly desirable. Although several computational methods have been developed for predicting the interactions between miRNA and target genes, there are substantial differences in the results achieved since most algorithms provide a large number of false positives. Accordingly, machine learning approaches are widely used to integrate predictions obtained from different tools. In this work, we adopt a method called multidimensional multiclass GP with multidimensional populations (M3GP), which relies on a genetic programming approach, to integrate and classify results from different miRNA-target prediction tools. The results are compared with those obtained with other classifiers, showing competitive accuracy. Since we aim to provide genome-wide predictions with M3GP and, considering the high number of miRNA-target interactions to test (also in different species), a parallel implementation of this algorithm is recommended. In this paper, we discuss the theoretical aspects of this algorithm and propose three different parallel implementations. We show that M3GP is highly parallelizable, it can be used to achieve genome-wide predictions, and its adoption provides great advantages when handling big datasets.

show abstract

Section: Resultssupporting

confidence: 76%

A Scalable Genetic Programming Approach to Integrate miRNA‐Target Predictions: Comparing Different Parallel Implementations of M3GP

et al. 2018

View full text Add to dashboard Cite

show abstract

“…The function set contained only the four basic arithmetic operators (+, −, ×, and ÷, protected against division by zero as in [60]), plus the Maximum (max) and Minimum (min) operators. Although there is a vast array of tunable parameters even in the most basic GP system, normally they do not substantially influence the outcome in terms of best fitness achieved [61].…”

Section: Genetic Programming and Supervised Learningmentioning

confidence: 99%

Evolving knowledge graph similarity for supervised learning in complex biomedical domains

Sousa¹,

Silva²,

Pesquita³

2020

BMC Bioinformatics

View full text Add to dashboard Cite

Background: In recent years, biomedical ontologies have become important for describing existing biological knowledge in the form of knowledge graphs. Data mining approaches that work with knowledge graphs have been proposed, but they are based on vector representations that do not capture the full underlying semantics. An alternative is to use machine learning approaches that explore semantic similarity. However, since ontologies can model multiple perspectives, semantic similarity computations for a given learning task need to be fine-tuned to account for this. Obtaining the best combination of semantic similarity aspects for each learning task is not trivial and typically depends on expert knowledge. Results: We have developed a novel approach, evoKGsim, that applies Genetic Programming over a set of semantic similarity features, each based on a semantic aspect of the data, to obtain the best combination for a given supervised learning task. The approach was evaluated on several benchmark datasets for protein-protein interaction prediction using the Gene Ontology as the knowledge graph to support semantic similarity, and it outperformed competing strategies, including manually selected combinations of semantic aspects emulating expert knowledge. evoKGsim was also able to learn species-agnostic models with different combinations of species for training and testing, effectively addressing the limitations of predicting protein-protein interactions for species with fewer known interactions. Conclusions: evoKGsim can overcome one of the limitations in knowledge graph-based semantic similarity applications: the need to expertly select which aspects should be taken into account for a given application. Applying this methodology to protein-protein interaction prediction proved successful, paving the way to broader applications.

show abstract

“…Evolutionary computation generates the first set of possible solutions and then produces iterative updates by the stochastic deletion of less desirable solutions and introduction of random minor changes. In a biological sense, each population of solutions is exposed to either natural or artificial selection and mutation [15].…”

Section: Evolutionary Computationmentioning

confidence: 99%

Untitled

2019

Review of Computer Engineering Research

View full text Add to dashboard Cite

Article HistoryKeywords Soft computing Software efficiency Software development Programming technology Soft computing components.This paper examines the methods of increasing software efficiency based on soft computing technology by analyzing the benefits and challenges of the components used in software development. The functions and features of machine learning, comprising neural networks, perceptron, support vector machines, and fuzzy logic are highlighted and explained, along with those of evolutionary computation (e.g., evolutionary and genetic algorithms) and probability. Recommendations are presented in conclusion. Contribution/Originality:This paper is one of only a few to investigate and analyze the problems of soft computing technologies. By reviewing the utilization of new estimation methodologies, it contributes to the existing literature and comparative analyses.

show abstract

Investigating the parameter space of evolutionary algorithms

Cited by 68 publications

References 29 publications

A Scalable Genetic Programming Approach to Integrate miRNA‐Target Predictions: Comparing Different Parallel Implementations of M3GP

A Scalable Genetic Programming Approach to Integrate miRNA‐Target Predictions: Comparing Different Parallel Implementations of M3GP

Evolving knowledge graph similarity for supervised learning in complex biomedical domains

Untitled

Contact Info

Product

Resources

About