Application of pool-based active learning in reducing the number of required response history analyses. (December 2020)

• Record Type: Journal Article
• Authors: Kiani, Jalal; Camp, Charles; Pezeshk, Shahram; Khoshnevis, Naeem
• Is Part Of: Computers & structures. Volume 241(2020)
• Publication Date: 2020-12
• Subjects: Machine learning -- Active learning -- Neural networks -- Response history analyses -- Fragility curve; RHA Response History Analysis -- GM Ground Motion -- AL Active Learning -- PL Passive Learning -- ANN Artificial Neural Network -- QBC Query by Committee; Structural engineering -- Data processing -- Periodicals; Electronic data processing -- Structures, Theory of -- Periodicals; 624.171
• DOI: 10.1016/j.compstruc.2020.106355
• Languages: English
• ISSNs: 0045-7949

Abstract

Highlights: Reduce the computational efforts for seismic risk quantification. Develop a novel method to reduce the need for so many response history analyses. An easy-to-implement machine learning based method for deriving fragility curves. Improve performance of machine learning tools in predicting the structural responses. Use pool-based active learning to find the most informative ground motion records. Abstract: A step by step method is presented for reducing the need for a large number of response history analyses (RHAs) in developing surrogates to predict the structural responses. These surrogates, which map ground motions features and characteristics of the structural systems into structural responses, are used in deriving fragility curves; and mostly are developed using machine learning algorithms. A machine learning algorithm, depending on the complexity of the model, requires a sufficient amount of training data to predict the outputs accurately. For complicated structural models, generating training data can be computationally demanding. Therefore, there is a need to generate the least amount of training data while preserving the accuracy of the prediction models. Towards this goal, a pool-based query-by-committee active learning (AL) algorithm is applied to choose probably the most informative data samples for surrogate training. A pool of unlabeled GM data samples is generated and a committee of artificial neural networks (ANNs) is defined to choose the smallest subset of the data that would be the most informative training data. The selected data samples meet informativeness, representativeness, and diversity criteria. The results of the applied case study show that implementing AL, can considerably reduce the size of required training data and consequently the amount of RHAs while improving the performance of the prediction models. Specifically, the findings demonstrate that AL improves the average F-scores of ANNs by about 10% and significantly reduces their variations, which indicates its stable behavior.