Architecture Optimization of a Non-Linear Autoregressive Neural Networks for Mackey-Glass Time Series Prediction Using Discrete Mycorrhiza Optimization Algorithm

Carreon-Ortiz, Hector; Valdez, Fevrier; Melín, Patricia; Castillo, Oscar

doi:10.3390/mi14010149

Cited by 5 publications

(3 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have previously applied the DMOA algorithm in the optimization of the architecture of a non-linear autoregressive neural network for Mackey-Glass time series prediction [69], and in this article in the adaptation of the parameters of the fuzzy systems T1FLS and IT2FLS. Additionally, hypothesis tests were carried out and the results obtained in both investigations were favorable for the DMOA optimization algorithm.…”

Section: Discussionmentioning

confidence: 99%

Comparative Study of Type-1 and Interval Type-2 Fuzzy Logic Systems in Parameter Adaptation for the Fuzzy Discrete Mycorrhiza Optimization Algorithm

2023

Self Cite

View full text Add to dashboard Cite

The Fuzzy Discrete Mycorrhiza Optimization (FDMOA) Algorithm is a new hybrid optimization method using the Discrete Mycorrhiza Optimization Algorithm (DMOA) in combination with type-1 or interval type-2 fuzzy logic system. In this new research, when using T1FLS, membership functions are defined by type-1 fuzzy sets, which allows for a more flexible and natural representation of uncertain and imprecise data. This approach has been successfully applied to several optimization problems, such as in feature selection, image segmentation, and data clustering. On the other hand, when DMOA is using IT2FLS, membership functions are represented by interval type-2 fuzzy sets, which allows for a more robust and accurate representation of uncertainty. This approach has been shown to handle higher levels of uncertainty and noise in the input data and has been successfully applied to various optimization problems, including control systems, pattern recognition, and decision-making. Both DMOA using T1FLS and DMOA using IT2FLS have shown better performance than the original DMOA algorithm in many applications. The combination of DMOA with fuzzy logic systems provides a powerful and flexible optimization framework that can be adapted to various problem domains. In addition, these techniques have the potential to more efficiently and effectively solve real-world problems.

show abstract

Section: Discussionmentioning

confidence: 99%

Comparative Study of Type-1 and Interval Type-2 Fuzzy Logic Systems in Parameter Adaptation for the Fuzzy Discrete Mycorrhiza Optimization Algorithm

2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…The models were trained for 20 epochs, and the best-performing show on the approval set was chosen for assessment on the test set. For reasonable comparison, the same exploratory setup was connected to the Vanilla RNN, GRU, Attention Mechanism, and Transformer-based models [11]. Hyperparameters, such as covered-up measure, input measure, and learning rate, were tuned based on approval set execution.…”

Section: Transformer-based Modelmentioning

confidence: 99%

Recurrent Neural Networks for Image Captioning: A Case Study with LSTM

Devi K, Manish Sharma, R. N. Patil, Anurag Shrivastava

2024

jes

View full text Add to dashboard Cite

This research investigates the viability of Long Short-Term Memory (LSTM) systems, a subtype of Recurrent Neural Networks (RNNs), for picture captioning. Leveraging the MS COCO dataset, the study compares the execution of LSTM-based RNNs with Vanilla RNN, Gated Recurrent Unit (GRU), consideration components, and transformer-based models. Experimental comes about to illustrate that the LSTM-based RNN shows competitive execution, accomplishing a BLEU-4 score of 0.72, a METEOR score of 0.68, and a CIDEr score of 2.1. The comparative investigation uncovers its prevalence over Vanilla RNN and GRU, highlighting its capability to capture long-range conditions inside successive picture information. Moreover, the study investigates the effect of consideration instruments and transformer designs, exhibiting their potential improvements in the context-aware caption era. The transformer-based show outflanks all other models, accomplishing a BLEU-4 score of 0.78, a METEOR score of 0.72, and a CIDEr score of 2.5. The findings give important bits of knowledge toward the creating scene of picture captioning strategy, which makes LSTM-based RNNs solid and productive approaches for capturing worldly groupings in visual substance. In achieving these, the study provides a framework for future developments in hybrid models and manufacturing processes that push boundaries of smart image perception and understanding

show abstract

“…Although they possess distinct advantages and challenges, both methods offer significant benefits in time series forecasting. Researchers worldwide are developing diverse artificial intelligence (AI) models to meet the growing need to predict future events accurately [5]. Neural Networks (NNs) and Fuzzy Systems (FS) are two types of artificial intelligence models that are increasingly utilized to make precise and dependable predictions in time series forecasting.…”

Section: Introductionmentioning

confidence: 99%

A Performance Comparison of Neural Networks and Fuzzy Systems for Time Series Forecasting

Woong

2024

JMC

View full text Add to dashboard Cite

Artificial neural networks and fuzzy structures have gained significant popularity in the last decade for time series forecasting. The objective is to conduct a performance comparison of various strategies to determine which ones are more effective for time series forecasting. The dataset provides instruction and evaluates forecasting models, utilizing artificial neural networks and fuzzy architectures. The observation evaluates the overall effectiveness of the forecasting models and the use of the root mean square error and means absolute error measures. This comparison analysis provides initial insights into the efficacy of artificial neural networks and fuzzy structures for predicting time series data. In predicting time series data, this study examines the precision of two renowned artificial intelligence systems, Neural Networks and Fuzzy structures. To evaluate the two algorithms, two distinct types of time series were utilized: a synthetic dataset consisting of 150 variables and a real-world dataset including 129 data points about monetary forecasts. The models' forecasting accuracy, training duration, and generalization abilities were compared. The findings validated that neural network surpassed fuzzy structures in all performance metrics when handling synthetic data. This research emphasizes the capabilities of artificial neural networks and fuzzy structures in addressing complicated forecasting problems. It demonstrates that both techniques may be utilized for predicting future time series values.

show abstract

Architecture Optimization of a Non-Linear Autoregressive Neural Networks for Mackey-Glass Time Series Prediction Using Discrete Mycorrhiza Optimization Algorithm

Cited by 5 publications

References 82 publications

Comparative Study of Type-1 and Interval Type-2 Fuzzy Logic Systems in Parameter Adaptation for the Fuzzy Discrete Mycorrhiza Optimization Algorithm

Comparative Study of Type-1 and Interval Type-2 Fuzzy Logic Systems in Parameter Adaptation for the Fuzzy Discrete Mycorrhiza Optimization Algorithm

Recurrent Neural Networks for Image Captioning: A Case Study with LSTM

A Performance Comparison of Neural Networks and Fuzzy Systems for Time Series Forecasting

Contact Info

Product

Resources

About