Kernel Corrector LSTM
This work addresses computational efficiency for time-series forecasting practitioners, but it is incremental as it builds on an existing method with a simpler kernel smoothing approach.
The paper tackles the problem of data quality issues in forecasting by proposing Kernel Corrector LSTM (KcLSTM), which reduces training time by 30-50% compared to Corrector LSTM while maintaining competitive forecasting accuracy with errors within 2-5% of baseline LSTM.
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.