Corrector LSTM: built-in training data correction for improved time-series forecasting

• Tipo: Artigo em Revista Científica Internacional
• Data: 2024

Autores

Baghoussi, Y (Autor) Outra; Carlos Soares (Autor) FEUP; João Mendes-Moreira (Autor) FEUP

Revista

Título: Neural Computing and Applications

Vol. 36

ISSN: 0941-0643

Editora: Springer Nature

Indexação

Scopus - 0 Citações

Outras Informações

ID Authenticus: P-010-J2C

DOI: 10.1007/s00521-024-09962-x

Abstract (EN): Traditional recurrent neural networks (RNNs) are essential for processing time-series data. However, they function as read-only models, lacking the ability to directly modify the data they learn from. In this study, we introduce the corrector long short-term memory (cLSTM), a Read & Write LSTM architecture that not only learns from the data but also dynamically adjusts it when necessary. The cLSTM model leverages two key components: (a) predicting LSTM¿s cell states using Seasonal Autoregressive Integrated Moving Average (SARIMA) and (b) refining the training data based on discrepancies between actual and forecasted cell states. Our empirical validation demonstrates that cLSTM surpasses read-only LSTM models in forecasting accuracy across the Numenta Anomaly Benchmark (NAB) and M4 Competition datasets. Additionally, cLSTM exhibits superior performance in anomaly detection compared to hierarchical temporal memory (HTM) models. © The Author(s) 2024.

Idioma: Inglês

Tipo (Avaliação Docente): Científica