An official website of the United States government
The US Drought Monitor (USDM) is a hallmark in real time drought monitoring and assessment as it was developed by multiple agencies to provide an accurate and timely assessment of drought conditions in the US on a weekly basis. The map is built based on multiple physical indicators as well as reported observations from local contributors before human analysts combine the information and produce the drought map using their best judgement. Since human subjectivity is included in the production of the USDM maps, it is not an entirely clear quantitative procedure for other entities to reproduce the maps. In this study, we developed a framework to automatically generate the maps through a machine learning approach by predicting the drought categories across the domain of study. A persistence model served as the baseline model for comparison in the framework. Three machine learning algorithms, logistic regression, random forests, and support vector machines, with four different groups of input data, which formed an overall of 12 different configurations, were used for the prediction of drought categories. Finally, all the configurations were evaluated against the baseline model to select the best performing option. The results showed that our proposed framework could reproduce the drought maps to a near-perfect level with the support vector machines algorithm and the group 4 data. The rest of the findings of this study can be highlighted as: 1) employing the past week drought data as a predictor in the models played an important role in achieving high prediction scores, 2) the nonlinear models, random forest, and support vector machines had a better overall performance compared to the logistic regression models, and 3) with borrowing the neighboring grid cells information, we could compensate the lack of training data in the grid cells with insufficient historical USDM data particularly for extreme and exceptional drought conditions.
more »
« less
This content will become publicly available on March 1, 2026
Applying Machine‐Learning Methods to Laser Acceleration of Protons: Lessons Learned From Synthetic Data
In this study, we consider three different machine‐learning methods—a three‐hidden‐layer neural network, support vector regression, and Gaussian process regression—and compare how well they can learn from a synthetic data set for proton acceleration in the Target Normal Sheath Acceleration regime. The synthetic data set was generated from a previously published theoretical model by Fuchs et al. 2005 that we modified. Once trained, these machine‐learning methods can assist with efforts to maximize the peak proton energy, or with the more general problem of configuring the laser system to produce a proton energy spectrum with desired characteristics. In our study, we focus on both the accuracy of the machine‐learning methods and the performance on one GPU including memory consumption. Although it is arguably the least sophisticated machine‐learning model we considered, support vector regression performed very well in our tests.
more »
« less
- Award ID(s):
- 2109222
- PAR ID:
- 10611577
- Publisher / Repository:
- Wiley Online
- Date Published:
- Journal Name:
- Contributions to Plasma Physics
- Volume:
- 65
- Issue:
- 3
- ISSN:
- 0863-1042
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Due to increasing volume of measurements in smart grids, surrogate based learning approaches for modeling the power grids are becoming popular. This paper uses regression based models to find the unknown state variables on power systems. Generally, to determine these states, nonlinear systems of power flow equations are solved iteratively. This study considers that the power flow problem can be modeled as an data driven type of a model. Then, the state variables, i.e., voltage magnitudes and phase angles are obtained using machine learning based approaches, namely, Extreme Learning Machine (ELM), Gaussian Process Regression (GPR), and Support Vector Regression (SVR). Several simulations are performed on the IEEE 14 and 30-Bus test systems to validate surrogate based learning based models. Moreover, input data was modified with noise to simulate measurement errors. Numerical results showed that all three models can find state variables reasonably well even with measurement noise.more » « less
-
Phase of flight (POF) prediction estimates the future state of aircraft along planned trajectories, allowing the prediction of potential conflicts as well as optimization of air space, controlled by the Federal Aviation Administration. In this paper, we present a study conducted to develop three different POF forecasting machine learning models and a statistical regression model using four-dimensional GPS and RADAR Track data from 57 flights provided by an En Route Computer System. The investigated machine learning models include Long Short-Term Memory Recurrent Neural Network (LSTM-RNN), Support Vector Machine (SVM), and Neural Ordinary Differential Equations (NODE). These were developed to forecast the horizontal and vertical POF of the current aircraft for the next time step. The results in this study indicate that LSTM-RNN models are more suitable for POF prediction than SVM and statistical regression models, with NODE being a promising model for future trajectory prediction research.more » « less
-
Le, Khanh N.Q. (Ed.)In current clinical settings, typically pain is measured by a patient’s self-reported information. This subjective pain assessment results in suboptimal treatment plans, over-prescription of opioids, and drug-seeking behavior among patients. In the present study, we explored automatic objective pain intensity estimation machine learning models using inputs from physiological sensors. This study uses BioVid Heat Pain Dataset. We extracted features from Electrodermal Activity (EDA), Electrocardiogram (ECG), Electromyogram (EMG) signals collected from study participants subjected to heat pain. We built different machine learning models, including Linear Regression, Support Vector Regression (SVR), Neural Networks and Extreme Gradient Boosting for continuous value pain intensity estimation. Then we identified the physiological sensor, feature set and machine learning model that give the best predictive performance. We found that EDA is the most information-rich sensor for continuous pain intensity prediction. A set of only 3 features from EDA signals using SVR model gave an average performance of 0.93 mean absolute error (MAE) and 1.16 root means square error (RMSE) for the subject-independent model and of 0.92 MAE and 1.13 RMSE for subject-dependent. The MAE achieved with signal-feature-model combination is less than 1 unit on 0 to 4 continues pain scale, which is smaller than the MAE achieved by the methods reported in the literature. These results demonstrate that it is possible to estimate pain intensity of a patient using a computationally inexpensive machine learning model with 3 statistical features from EDA signal which can be collected from a wrist biosensor. This method paves a way to developing a wearable pain measurement device.more » « less
-
Abstract Water sustainability in the built environment requires an accurate estimation of residential water end uses (e.g., showers, toilets, faucets, etc.). In this study, we evaluate the performance of four models (Random Forest, RF; Support Vector Machines, SVM; Logistic Regression, Log‐reg; and Neural Networks, NN) for residential water end‐use classification using actual (measured) and synthetic labeled data sets. We generated synthetic labeled data using Conditional Tabular Generative Adversarial Networks. We then utilized grid search to train each model on their respective optimized hyperparameters. The RF model exhibited the best model performance overall, while the Log‐reg model had the shortest execution times under different balanced and imbalanced (based on number of events per class) synthetic data scenarios, demonstrating a computationally efficient alternative for RF for specific end uses. The NN model exhibited high performance with the tradeoff of longer execution times compared to the other classification models. In the balanced data set scenario, all models achieved closely aligned F1‐scores, ranging from 0.83 to 0.90. However, when faced with imbalanced data reflective of actual conditions, both the SVM and Log‐reg models showed inferior performance compared to the RF and NN models. Overall, we concluded that decision tree‐based models emerge as the optimal choice for classification tasks in the context of water end‐use data. Our study advances residential smart water metering systems through creating synthetic labeled end‐use data and providing insight into the strengths and weaknesses of various supervised machine learning classifiers for end‐use identification.more » « less
