<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
Journal article(2022)
-
D. Eeltink, H. Branger, C. Luneau, Y. He, A. Chabchoub, J. Kasparian, T.S. van den Bremer, T. P. Sapsis
Wave breaking is the main mechanism that dissipates energy input into ocean waves by wind and transferred across the spectrum by nonlinearity. It determines the properties of a sea state and plays a crucial role in ocean-atmosphere interaction, ocean pollution, and rogue waves. Owing to its turbulent nature, wave breaking remains too computationally demanding to solve using direct numerical simulations except in simple, short-duration circumstances. To overcome this challenge, we present a blended machine learning framework in which a physics-based nonlinear evolution model for deep-water, non-breaking waves and a recurrent neural network are combined to predict the evolution of breaking waves. We use wave tank measurements rather than simulations to provide training data and use a long short-term memory neural network to apply a finite-domain correction to the evolution model. Our blended machine learning framework gives excellent predictions of breaking and its effects on wave evolution, including for external data.
...
Wave breaking is the main mechanism that dissipates energy input into ocean waves by wind and transferred across the spectrum by nonlinearity. It determines the properties of a sea state and plays a crucial role in ocean-atmosphere interaction, ocean pollution, and rogue waves. Owing to its turbulent nature, wave breaking remains too computationally demanding to solve using direct numerical simulations except in simple, short-duration circumstances. To overcome this challenge, we present a blended machine learning framework in which a physics-based nonlinear evolution model for deep-water, non-breaking waves and a recurrent neural network are combined to predict the evolution of breaking waves. We use wave tank measurements rather than simulations to provide training data and use a long short-term memory neural network to apply a finite-domain correction to the evolution model. Our blended machine learning framework gives excellent predictions of breaking and its effects on wave evolution, including for external data.
Journal article(2020)
-
X. Li, Y. He, F. Fioranelli, X. Jing, A. Yarovoy, Y. Yang
The performance of deep learning (DL) algorithms for radar-based human motion recognition (HMR) is hindered by the diversity and volume of the available training data. In this article, to tackle the issue of insufficient training data for HMR, we propose an instance-based transfer learning (ITL) method with limited radar micro-Doppler (MD) signatures, alleviating the burden of collecting and annotating a large number of radar samples. ITL is a unique algorithm that consists of three interconnected parts, including DL model pretraining, correlated source data selection, and adaptive collaborative fine-tuning (FT). Any of the three components cannot be excluded; otherwise, the performance of the entire algorithm decreases. The experiments with a radar data set of six human motions show that ITL achieves state-of-the-art performance for HMR with limited training samples, outperforming several existing transfer learning approaches. Especially, when there are only 100 samples per person per class, ITL yields an F1 score of 96.7%. Last but not least, ITL is more generalized to human motion differences. Though adapted to recognize the persons' motions in a small-scale target data set, ITL can also classify the persons' motion data used for pretraining, achieving up to 11.0% F1 score enhancement over the conventional FT method.
...
The performance of deep learning (DL) algorithms for radar-based human motion recognition (HMR) is hindered by the diversity and volume of the available training data. In this article, to tackle the issue of insufficient training data for HMR, we propose an instance-based transfer learning (ITL) method with limited radar micro-Doppler (MD) signatures, alleviating the burden of collecting and annotating a large number of radar samples. ITL is a unique algorithm that consists of three interconnected parts, including DL model pretraining, correlated source data selection, and adaptive collaborative fine-tuning (FT). Any of the three components cannot be excluded; otherwise, the performance of the entire algorithm decreases. The experiments with a radar data set of six human motions show that ITL achieves state-of-the-art performance for HMR with limited training samples, outperforming several existing transfer learning approaches. Especially, when there are only 100 samples per person per class, ITL yields an F1 score of 96.7%. Last but not least, ITL is more generalized to human motion differences. Though adapted to recognize the persons' motions in a small-scale target data set, ITL can also classify the persons' motion data used for pretraining, achieving up to 11.0% F1 score enhancement over the conventional FT method.
