Hyperparameter Optimization for Deep NeuralNetwork Models: A Comprehensive Study onMethods and Techniques DOI Creative Commons
Sunita Roy, Ranjan Mehera, Rajat Kumar Pal

et al.

Research Square (Research Square), Journal Year: 2023, Volume and Issue: unknown

Published: June 23, 2023

Abstract Advancements in computing and storage technologies have significantly contributed to the adoption of deep learning (DL)-based models among machinelearning (ML) experts. Although a generic model can be used search fora near-optimal solution any problem domain, what makes these DL modelscontext-sensitive is combination training data hyperparameters. Due lack inherent explainability HyperparameterOptimization (HPO) or tuning specific each art,science, experience. In this article, we explored various existing methods ways identify optimal set values for hyperparameters specificto along with techniques realize those real-lifesituations. The article also includes detailed comparative study variousstate-of-the-art HPO using Keras Tuner toolkit highlights observations describing how performance improvedby applying techniques.

Language: Английский

Application of Meta-Heuristic Algorithms for Training Neural Networks and Deep Learning Architectures: A Comprehensive Review DOI Open Access
Mehrdad Kaveh, Mohammad Saadi Mesgari

Neural Processing Letters, Journal Year: 2022, Volume and Issue: 55(4), P. 4519 - 4622

Published: Oct. 31, 2022

Language: Английский

Citations

137
Sparrow search algorithm with adaptive t distribution for multi-objective low-carbon multimodal transportation planning problem with fuzzy demand and fuzzy time DOI

Huizhen Zhang,

Qin Huang, Liang Ma

et al.

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 238, P. 122042 - 122042

Published: Oct. 12, 2023

Language: Английский

Citations

27
Optimizing RNNs for EMG Signal Classification: A Novel Strategy Using Grey Wolf Optimization DOI Creative Commons
Marcos Avilés, José M. Álvarez-Alvarado, J.B. Robles-Ocampo

et al.

Bioengineering, Journal Year: 2024, Volume and Issue: 11(1), P. 77 - 77

Published: Jan. 13, 2024

Accurate classification of electromyographic (EMG) signals is vital in biomedical applications. This study evaluates different architectures recurrent neural networks for the EMG associated with five movements right upper extremity. A Butterworth filter was implemented signal preprocessing, followed by segmentation into 250 ms windows, an overlap 190 ms. The resulting dataset divided training, validation, and testing subsets. Grey Wolf Optimization algorithm applied to gated unit (GRU), long short-term memory (LSTM) architectures, bidirectional networks. In parallel, a performance comparison support vector machines (SVMs) performed. results obtained first experimental phase revealed that all RNN evaluated reached 100% accuracy, standing above 93% achieved SVM. Regarding speed, LSTM ranked as fastest architecture, recording time 0.12 ms, GRU 0.134 Bidirectional showed response 0.2 while SVM had longest at 2.7 second phase, slight decrease accuracy models observed, 98.46% LSTM, 96.38% GRU, 97.63% network. findings this highlight effectiveness speed task.

Language: Английский

Citations

9
Storm-time ionospheric model over Yunnan-Sichuan area of China based on the SSA-ConvLSTM-BiLSTM algorithm DOI
Wang Li, H. Zhu,

Fangsong Yang

et al.

GPS Solutions, Journal Year: 2025, Volume and Issue: 29(2)

Published: March 10, 2025

Language: Английский

Citations

1
Novel hybrid success history intelligent optimizer with Gaussian transformation: application in CNN hyperparameter tuning DOI
Hussam N. Fakhouri, Sadi Alawadi, Feras M. Awaysheh

et al.

Cluster Computing, Journal Year: 2023, Volume and Issue: 27(3), P. 3717 - 3739

Published: Nov. 6, 2023

Language: Английский

Citations

16
Phase prediction of high-entropy alloys based on machine learning and an improved information fusion approach DOI Creative Commons
Cun Chen,

Xiaoli Han,

Yong Zhang

et al.

Computational Materials Science, Journal Year: 2024, Volume and Issue: 239, P. 112976 - 112976

Published: March 29, 2024

Language: Английский

Citations

5
Hyperparameter optimization for deep neural network models: a comprehensive study on methods and techniques DOI
Sunita Roy, Ranjan Mehera, Rajat Kumar Pal

et al.

Innovations in Systems and Software Engineering, Journal Year: 2023, Volume and Issue: unknown

Published: Oct. 7, 2023

Language: Английский

Citations

10
An enhanced sparrow search swarm optimizer via multi-strategies for high-dimensional optimization problems DOI
Shuang Liang, Minghao Yin, Geng Sun

et al.

Swarm and Evolutionary Computation, Journal Year: 2024, Volume and Issue: 88, P. 101603 - 101603

Published: May 18, 2024

Language: Английский

Citations

3
Modeling Short-Term Drought for SPEI in Mainland China Using the XGBoost Model DOI Creative Commons

Fanchao Zeng,

Q. Gao, Lifeng Wu

et al.

Atmosphere, Journal Year: 2025, Volume and Issue: 16(4), P. 419 - 419

Published: April 4, 2025

Accurate drought prediction is crucial for optimizing water resource allocation, safeguarding agricultural productivity, and maintaining ecosystem stability. This study develops a methodological framework short-term forecasting using SPEI time series (1979–2020) evaluates three predictive models: (1) baseline XGBoost model (XGBoost1), (2) feature-optimized variant incorporating Pearson correlation analysis (XGBoost2), (3) an enhanced CPSO-XGBoost integrating hybrid particle swarm optimization with dual mechanisms of binary feature selection parameter tuning. Key findings reveal spatiotemporal patterns: temporal-scale dependencies show all models exhibit limited capability at SPEI-1 (R2: 0.32–0.41, RMSE: 0.68–0.79) but achieve progressive accuracy improvement, peaking SPEI-12 where attains optimal performance 0.85–0.90, 0.33–0.43) 18.7–23.4% error reduction versus baselines. Regionally, humid zones (South China/Central-Southern) demonstrate peak (R2 ≈ 0.90, RMSE < 0.35), while arid regions (Northwest Desert/Qinghai-Tibet Plateau) dramatic improvement from 0.35, > 1.0) to 0.85, 52%). Multivariate probability density confirms the model’s robustness through capture nonlinear atmospheric-land interactions reduced parameterization uncertainties via intelligence optimization. The CPSO-XGBoost’s superiority stems synergistic optimization: enhances input relevance adaptive tuning improves computational efficiency, collectively addressing climate variability challenges across diverse terrains. These establish advanced early warning systems, providing critical support climate-resilient management risk mitigation spatiotemporally predictions.

Language: Английский

Citations

0
Online Prediction of Concrete Temperature During the Construction of an Arch Dam Based on a Sparrow Search Algorithm–Incremental Support Vector Regression Model DOI Creative Commons
Yihong Zhou, Deng Yu, Fang Wang

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(9), P. 5053 - 5053

Published: May 1, 2025

The accurate prediction of concrete temperature during arch dam construction is essential for crack prevention. internal the poured blocks influenced by dynamic factors such as material properties, age, heat dissipation conditions, and control measures, which are highly time-varying. Conventional models, rely on offline data training, struggle to capture these time-varying dynamics, resulting in insufficient accuracy. To overcome limitations, this study constructed a sparrow search algorithm–incremental support vector regression (SSA-ISVR) model online prediction. First, SSA was employed optimize penalty kernel coefficients ISVR algorithm, minimizing errors between predicted measured temperatures establish pretrained initial model. Second, untrained samples were dynamically monitored incorporated using Karush–Kuhn–Tucker (KKT) conditions identify unlearned information, prompting updates. Additionally, redundant removed based sample similarity error-driven criteria enhance training efficiency. Finally, model’s accuracy reliability validated through actual case studies compared LSTM, BP, models. results indicate that SSA-ISVR outperforms aforementioned effectively capturing changes accurately predicting variations, with mean absolute error 0.14 °C.

Language: Английский

Citations

0