Abstract
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.

Abstract
Forecasting methods are affected by data quality issues in two ways: 1. they are hard to predict, and 2. they may affect the model negatively when it is updated with new data. The latter issue is usually addressed by pre-processing the data to remove those issues. An alternative approach has recently been proposed, Corrector LSTM (cLSTM), which is a Read & Write Machine Learning (RW-ML) algorithm that changes the data while learning to improve its predictions. Despite promising results being reported, cLSTM is computationally expensive, as it uses a meta-learner to monitor the hidden states of the LSTM. We propose a new RW-ML algorithm, Kernel Corrector LSTM (KcLSTM), that replaces the meta-learner of cLSTM with a simpler method: Kernel Smoothing. We empirically evaluate the forecasting accuracy and the training time of the new algorithm and compare it with cLSTM and LSTM. Results indicate that it is able to decrease the training time while maintaining a competitive forecasting accuracy.

Abstract
Machine learning models are widely used in time series forecasting. One way to reduce its computational cost and increase its efficiency is to select only the relevant exogenous features to be fed into the model. With this intention, a study on the feature selection methods: Pearson correlation coefficient, Boruta, Boruta-Shap, IMV-LSTM, and LIME is performed. A new method focused on interpretability, SHAP-LSTM, is proposed, using a deep learning model training process as part of a feature selection algorithm. The methods were compared in 2 different datasets showing comparable results with lesser computational cost when compared with the use of all features. In all datasets, SHAP-LSTM showed competitive results, having comparatively better results on the data with a higher presence of scarce occurring categorical features.

Abstract