Kernel Corrector LSTM

TL;DR

This paper introduces Kernel Corrector LSTM (KcLSTM), a new forecasting algorithm that reduces training time by replacing the meta-learner in Corrector LSTM with Kernel Smoothing, maintaining accuracy.

Contribution

It proposes a simplified RW-ML algorithm, KcLSTM, replacing the meta-learner with Kernel Smoothing to improve efficiency while preserving forecasting performance.

Findings

KcLSTM reduces training time compared to cLSTM.

KcLSTM maintains competitive forecasting accuracy.

Empirical evaluation shows effectiveness of the proposed method.

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.