Search Results
You are looking at 1 - 5 of 5 items for
- Author or Editor: Philippe Tissot x
- Refine by Access: All Content x
Abstract
Resampling methods such as cross validation or bootstrap are often employed to estimate the uncertainty in a loss function due to sampling variability, usually for the purpose of model selection. In models that require nonlinear optimization, however, the existence of local minima in the loss function landscape introduces an additional source of variability that is confounded with sampling variability. In other words, some portion of the variability in the loss function across different resamples is due to local minima. Given that statistically sound model selection is based on an examination of variance, it is important to disentangle these two sources of variability. To that end, a methodology is developed for estimating each, specifically in the context of K-fold cross validation, and neural networks (NN) whose training leads to different local minima. Random effects models are used to estimate the two variance components—that due to sampling and that due to local minima. The results are examined as a function of the number of hidden nodes, and the variance of the initial weights, with the latter controlling the “depth” of local minima. The main goal of the methodology is to increase statistical power in model selection and/or model comparison. Using both simulated and realistic data, it is shown that the two sources of variability can be comparable, casting doubt on model selection methods that ignore the variability due to local minima. Furthermore, the methodology is sufficiently flexible so as to allow assessment of the effect of other/any NN parameters on variability.
Abstract
Resampling methods such as cross validation or bootstrap are often employed to estimate the uncertainty in a loss function due to sampling variability, usually for the purpose of model selection. In models that require nonlinear optimization, however, the existence of local minima in the loss function landscape introduces an additional source of variability that is confounded with sampling variability. In other words, some portion of the variability in the loss function across different resamples is due to local minima. Given that statistically sound model selection is based on an examination of variance, it is important to disentangle these two sources of variability. To that end, a methodology is developed for estimating each, specifically in the context of K-fold cross validation, and neural networks (NN) whose training leads to different local minima. Random effects models are used to estimate the two variance components—that due to sampling and that due to local minima. The results are examined as a function of the number of hidden nodes, and the variance of the initial weights, with the latter controlling the “depth” of local minima. The main goal of the methodology is to increase statistical power in model selection and/or model comparison. Using both simulated and realistic data, it is shown that the two sources of variability can be comparable, casting doubt on model selection methods that ignore the variability due to local minima. Furthermore, the methodology is sufficiently flexible so as to allow assessment of the effect of other/any NN parameters on variability.
An Approach to Approximate Wave Height from Acoustic Tide GaugesAbstract
The demand for nearshore wave observations is increasing due to spatial gaps and the importance of observations for accurate models and better understanding of inundation processes. Here, we show how water level (WL) standard deviation (sigma, σ) measurements at three acoustic NOAA tide gauges that utilize an Aquatrak sensor [Duck, North Carolina, Bob Hall Pier (BHP) in Corpus Christi, Texas, and Lake Worth, Florida] can be used as a proxy for significant wave height (H m0). Sigma-derived H m0 is calibrated to best fit nearby wave observations and error is quantified through RMSE, normalized RMSE (NRMSE), bias, and a scatter index. At Duck and Lake Worth, a quadratic fit of sigma to nearby wave observations results in a R 2 of 0.97 and 0.83, RMSE of 0.11 and 0.11 m, and NRMSE of 0.09 and 0.22, respectively. A linear fit between BHP sigma and H m0 is best, resulting in R 2 0.62, RMSE of 0.22, and NRMSE of 0.26. Regression fits deviate across NOAA stations and from the classic relationship of H m0 = 4σ, indicating H m0 cannot be accurately estimated with this approach at these Aquatrak sites. The dynamic water level (DWL = still WL ± 2σ) is calculated over the historic time series showing climatological and seasonal trends in the stations’ daily maximums. The historical DWL and sigma wave proxy could be calculated for many NOAA tide gauges dating back to 1996. These historical wave observations can be used to fill observational spatial gaps, validate models, and improve understanding of wave climates.
Significance Statement
There is a large spatial gap in nearshore real-time observational wave data that can provide critical information to researchers and resource managers regarding inundation and erosion, help validate coastal hydrodynamic models, and provide the maritime community with products that help ensure navigational safety. This study utilizes existing infrastructure to help fill the demand for nearshore wave observations by deriving a proxy for wave height at three sites. This work shows spatial variability in the regression fits across the sites, which should be explored at more stations in future work. Multidecadal length time series were also used at the sites to investigate climatological and seasonal trends that provide insight into wave climates and wave driven processes important for coastal flooding.
Abstract
The demand for nearshore wave observations is increasing due to spatial gaps and the importance of observations for accurate models and better understanding of inundation processes. Here, we show how water level (WL) standard deviation (sigma, σ) measurements at three acoustic NOAA tide gauges that utilize an Aquatrak sensor [Duck, North Carolina, Bob Hall Pier (BHP) in Corpus Christi, Texas, and Lake Worth, Florida] can be used as a proxy for significant wave height (H m0). Sigma-derived H m0 is calibrated to best fit nearby wave observations and error is quantified through RMSE, normalized RMSE (NRMSE), bias, and a scatter index. At Duck and Lake Worth, a quadratic fit of sigma to nearby wave observations results in a R 2 of 0.97 and 0.83, RMSE of 0.11 and 0.11 m, and NRMSE of 0.09 and 0.22, respectively. A linear fit between BHP sigma and H m0 is best, resulting in R 2 0.62, RMSE of 0.22, and NRMSE of 0.26. Regression fits deviate across NOAA stations and from the classic relationship of H m0 = 4σ, indicating H m0 cannot be accurately estimated with this approach at these Aquatrak sites. The dynamic water level (DWL = still WL ± 2σ) is calculated over the historic time series showing climatological and seasonal trends in the stations’ daily maximums. The historical DWL and sigma wave proxy could be calculated for many NOAA tide gauges dating back to 1996. These historical wave observations can be used to fill observational spatial gaps, validate models, and improve understanding of wave climates.
Significance Statement
There is a large spatial gap in nearshore real-time observational wave data that can provide critical information to researchers and resource managers regarding inundation and erosion, help validate coastal hydrodynamic models, and provide the maritime community with products that help ensure navigational safety. This study utilizes existing infrastructure to help fill the demand for nearshore wave observations by deriving a proxy for wave height at three sites. This work shows spatial variability in the regression fits across the sites, which should be explored at more stations in future work. Multidecadal length time series were also used at the sites to investigate climatological and seasonal trends that provide insight into wave climates and wave driven processes important for coastal flooding.
Abstract
Artificial intelligence (AI) and machine learning (ML) have become important tools for environmental scientists and engineers, both in research and in applications. Although these methods have become quite popular in recent years, they are not new. The use of AI methods began in the 1950s and environmental scientists were adopting them by the 1980s. Although an “AI winter” temporarily slowed the growth, a more recent resurgence has brought it back with gusto. This paper tells the story of the evolution of AI in the field through the lens of the AMS Committee on Artificial Intelligence Applications to Environmental Science. The environmental sciences possess a host of problems amenable to advancement by intelligent techniques. We review a few of the early applications along with the ML methods of the time and how their progression has impacted these sciences. While AI methods have changed from expert systems in the 1980s to neural networks and other data-driven methods, and more recently deep learning, the environmental problems tackled have remained similar. We discuss the types of applications that have shown some of the biggest advances due to AI usage and how they have evolved over the past decades, including topics in weather forecasting, probabilistic prediction, climate estimation, optimization problems, image processing, and improving forecasting models. We finish with a look at where AI as employed in environmental science appears to be headed and some thoughts on how it might be best blended with physical/dynamical modeling approaches to further advance our science.
Abstract
Artificial intelligence (AI) and machine learning (ML) have become important tools for environmental scientists and engineers, both in research and in applications. Although these methods have become quite popular in recent years, they are not new. The use of AI methods began in the 1950s and environmental scientists were adopting them by the 1980s. Although an “AI winter” temporarily slowed the growth, a more recent resurgence has brought it back with gusto. This paper tells the story of the evolution of AI in the field through the lens of the AMS Committee on Artificial Intelligence Applications to Environmental Science. The environmental sciences possess a host of problems amenable to advancement by intelligent techniques. We review a few of the early applications along with the ML methods of the time and how their progression has impacted these sciences. While AI methods have changed from expert systems in the 1980s to neural networks and other data-driven methods, and more recently deep learning, the environmental problems tackled have remained similar. We discuss the types of applications that have shown some of the biggest advances due to AI usage and how they have evolved over the past decades, including topics in weather forecasting, probabilistic prediction, climate estimation, optimization problems, image processing, and improving forecasting models. We finish with a look at where AI as employed in environmental science appears to be headed and some thoughts on how it might be best blended with physical/dynamical modeling approaches to further advance our science.
Abstract
We introduce the National Science Foundation (NSF) AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES). This AI institute was funded in 2020 as part of a new initiative from the NSF to advance foundational AI research across a wide variety of domains. To date AI2ES is the only NSF AI institute focusing on environmental science applications. Our institute focuses on developing trustworthy AI methods for weather, climate, and coastal hazards. The AI methods will revolutionize our understanding and prediction of high-impact atmospheric and ocean science phenomena and will be utilized by diverse, professional user groups to reduce risks to society. In addition, we are creating novel educational paths, including a new degree program at a community college serving underrepresented minorities, to improve workforce diversity for both AI and environmental science.
Abstract
We introduce the National Science Foundation (NSF) AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES). This AI institute was funded in 2020 as part of a new initiative from the NSF to advance foundational AI research across a wide variety of domains. To date AI2ES is the only NSF AI institute focusing on environmental science applications. Our institute focuses on developing trustworthy AI methods for weather, climate, and coastal hazards. The AI methods will revolutionize our understanding and prediction of high-impact atmospheric and ocean science phenomena and will be utilized by diverse, professional user groups to reduce risks to society. In addition, we are creating novel educational paths, including a new degree program at a community college serving underrepresented minorities, to improve workforce diversity for both AI and environmental science.
Abstract
Promising new opportunities to apply artificial intelligence (AI) to the Earth and environmental sciences are identified, informed by an overview of current efforts in the community. Community input was collected at the first National Oceanic and Atmospheric Administration (NOAA) workshop on “Leveraging AI in the Exploitation of Satellite Earth Observations and Numerical Weather Prediction” held in April 2019. This workshop brought together over 400 scientists, program managers, and leaders from the public, academic, and private sectors in order to enable experts involved in the development and adaptation of AI tools and applications to meet and exchange experiences with NOAA experts. Paths are described to actualize the potential of AI to better exploit the massive volumes of environmental data from satellite and in situ sources that are critical for numerical weather prediction (NWP) and other Earth and environmental science applications. The main lessons communicated from community input via active workshop discussions and polling are reported. Finally, recommendations are presented for both scientists and decision-makers to address some of the challenges facing the adoption of AI across all Earth science.
Abstract
Promising new opportunities to apply artificial intelligence (AI) to the Earth and environmental sciences are identified, informed by an overview of current efforts in the community. Community input was collected at the first National Oceanic and Atmospheric Administration (NOAA) workshop on “Leveraging AI in the Exploitation of Satellite Earth Observations and Numerical Weather Prediction” held in April 2019. This workshop brought together over 400 scientists, program managers, and leaders from the public, academic, and private sectors in order to enable experts involved in the development and adaptation of AI tools and applications to meet and exchange experiences with NOAA experts. Paths are described to actualize the potential of AI to better exploit the massive volumes of environmental data from satellite and in situ sources that are critical for numerical weather prediction (NWP) and other Earth and environmental science applications. The main lessons communicated from community input via active workshop discussions and polling are reported. Finally, recommendations are presented for both scientists and decision-makers to address some of the challenges facing the adoption of AI across all Earth science.
