Reservoir Computing in Epidemiological Forecasting: Predicting Chicken Pox incidence

Kumar, Kaushal

doi:10.1101/2023.04.24.23289018

Cited by 4 publications

(2 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2019, Himan et al introduced the Seagull Optimization Algorithm (SOA), inspired by the migration and aggressive behavior observed in seagull groups [5]. This algorithm is designed to simulate the strategic actions of seagulls, allowing them to navigate and avoid collisions, locate optimal positions, and efficiently attack during migration and foraging, all with the ultimate goal of discovering optimal solutions.…”

Section: Seagull Optimization Algorithmmentioning

confidence: 99%

Research on the purification mechanism of heavy metal pollution by biochar composites driven by degree learning

Dai,

Jia,

Zhan

et al. 2024

E3S Web Conf.

View full text Add to dashboard Cite

This paper proposes an innovative approach by integrating deep learning technology, specifically employing the GRU recurrent neural network model based on the Seagull optimization algorithm, to enhance the accuracy of predicting biochar performance. The Seagull optimization algorithm, inspired by seagull predatory behavior, is adept at efficiently identifying optimal model parameters, thereby improving the model’s generalization ability and robustness. The GRU recurrent neural network, designed for sequence data processing, proves to be instrumental in capturing dynamic and nonlinear interactions between biochar and heavy metals. This, in turn, contributes to heightened prediction accuracy and model interpretability. The article unfolds in a structured manner, beginning with an introduction to the biochar preparation method and its characteristics. It then delves into an analysis of the sources and hazards of heavy metal pollution. Following this, the paper explains the principles and advantages of deep learning technology, providing a comprehensive foundation for the subsequent discussion. The construction and verification process of the proposed model is then detailed, concluding with the presentation of experimental results and in-depth analysis. In essence, this research introduces a pioneering idea and methodology for optimizing biochar design and effectively controlling heavy metal pollution, presenting a fresh perspective on addressing these environmental challenges.

show abstract

Section: Seagull Optimization Algorithmmentioning

confidence: 99%

Research on the purification mechanism of heavy metal pollution by biochar composites driven by degree learning

Dai,

Jia,

Zhan

et al. 2024

E3S Web Conf.

View full text Add to dashboard Cite

show abstract

“…A common choice for the activation function is a hyperbolic tangent, f(x) = tanh(x) [31], but other choices are possible [24] and will be explored here. RCs can be used for recognition or prediction tasks [11,[31][32][33]. For prediction, a commonly used benchmark is the chaotic Lorenz system [34]:…”

Section: Forecasting With Rcsmentioning

confidence: 99%

Tuning the activation function to optimize the forecast horizon of a reservoir computer

Hurley,

Restrepo,

Shaheen

2024

J. Phys. Complex.

View full text Add to dashboard Cite

Reservoir computing is a machine learning framework where the readouts from a nonlinear system (the reservoir) are trained so that the output from the reservoir, when forced with an input signal, reproduces a desired output signal. A common implementation of reservoir computers is to use a recurrent neural network as the reservoir. The design of this network can have significant effects on the performance of the reservoir computer. In this paper we study the effect of the node activation function on the ability of reservoir computers to learn and predict chaotic time series. We find that the Forecast Horizon (FH), the time during which the reservoir’s predictions remain accurate, can vary by an order of magnitude across a set of 16 activation functions used in machine learning. By using different functions from this set, and by modifying their parameters, we explore whether the entropy of node activation levels or the curvature of the activation functions determine the predictive ability of the reservoirs. We find that the FH is low when the activation function is used in a region where it has low curvature, and a positive correlation between curvature and FH. For the activation functions studied we find that the largest FH generally occurs at intermediate levels of the entropy of node activation levels. Our results show that the performance of reservoir computers is very sensitive to the activation function shape. Therefore, modifying this shape in hyperparameter optimization algorithms can lead to improvements in reservoir computer performance.

show abstract

Forecasting Crude Oil Prices Using Reservoir Computing Models

Kumar

2024

Comput Econ

View full text Add to dashboard Cite

Accurate forecasting of crude oil prices is crucial for informed financial decision-making. This study presents a cutting-edge Reservoir Computing (RC) model specifically designed for precise crude oil price predictions, outperforming traditional methods such as ARIMA, LSTM, and GRU. Using daily closing prices from major indices spanning January 2010 to December 2023, we conducted a thorough evaluation. The RC model consistently demonstrates superior accuracy and computational efficiency. Quantitative metrics reveal the RC model’s dominance with a Mean Absolute Error (MAE) of 0.0094, Mean Squared Error (MSE) of 0.00035, Root Mean Squared Error (RMSE) of 0.0196, and a notably low Mean Absolute Percentage Error (MAPE) of $$1.450\%$$ 1.450 % . Additionally, the RC model’s runtime of 1.11 s underscores its computational efficiency, far surpassing ARIMA (493.22 s), LSTM (423.55 s), and GRU (15.73 s). During periods of economic disruption, such as the COVID-19 lockdowns, the RC model effectively captured sharp price fluctuations, highlighting its robust forecasting capability. These findings emphasize the RC model’s potential as a reliable tool for enhancing decision-making processes in the dynamic energy market, particularly for real-time applications such as infectious disease case count forecasting. This study advocates for the broader adoption of Reservoir Computing models to improve predictive accuracy and operational efficiency in energy economics.

show abstract

Reservoir Computing in Epidemiological Forecasting: Predicting Chicken Pox incidence

Cited by 4 publications

References 25 publications

Research on the purification mechanism of heavy metal pollution by biochar composites driven by degree learning

Research on the purification mechanism of heavy metal pollution by biochar composites driven by degree learning

Tuning the activation function to optimize the forecast horizon of a reservoir computer

Forecasting Crude Oil Prices Using Reservoir Computing Models

Contact Info

Product

Resources

About