Time Series Forecasting Based on Augmented Long Short-Term Memory

By Daniel Hsu. 2017

45 citations recorded by the source corpus.

In this paper, we use recurrent autoencoder model to predict the time series in single and multiple steps ahead. Previous prediction methods, such as recurrent neural network (RNN) and deep belief network (DBN) models, cannot learn long term dependencies. And conventional long short-term memory (LSTM) model doesn't remember recent inputs. Combining LSTM and autoencoder (AE), the proposed model can capture long-term dependencies across data points and uses features extracted from recent observations for augmenting LSTM at the same time. Based on comprehensive experiments, we show that the proposed methods significantly improves the state-of-art performance on chaotic time series benchmark and also has better performance on real-world data. Both single-output and multiple-output predictions are investigated.

Subfields: Artificial Intelligence, Building and Construction, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jan Gasthaus
2. Tim Januschowski
3. Bernie Wang
4. Danielle C. Maddix
5. Laurent Callot
6. Syama Sundar Rangapuram
7. Michael Bohlke-Schneider
8. David Salinas

In Building and Construction

1. Cornelis W. Oosterlee
2. Sander M. Bohte
3. Anastasia Borovykh

In Signal Processing

1. Bernie Wang
2. Tim Januschowski
3. Jan Gasthaus
4. Syama Sundar Rangapuram
5. Danielle C. Maddix
6. Laurent Callot
7. David Salinas
8. Michael Bohlke-Schneider

Citations in the corpus

• Conditional Time Series Forecasting with Convolutional Neural Networks (2017, background): “Currently, recurrent neural networks (RNNs), and in particular the long-short term memory unit (LSTM) [12], [5], are the state-of-the-art in time series forecasting, see also [14] and in particular [7] for financial forecasting results.”
• Dilated convolutional neural networks for time series forecasting (2018, background): “Currently, recurrent neural networks (RNNs), and in particular the long-short term memory unit (LSTM) [12], [5], are the state-of-the-art in time series forecasting, see also [14] and in particular [7] for financial forecasting results.”
• A Deep Neural Network Model for Short-Term Load Forecast Based on Long Short-Term Memory Network and Convolutional Neural Network (2018, background): “It is particularly suitable for modeling sequence problems by operating on input information as well as a trace of previously acquired information due to recurrent connections [32].”
• Deep Learning for Time Series Forecasting: Tutorial and Literature Survey (2020, methodology): “A-LSTM [82] RNN + MLP Point 1 `2, `2 regularizer RMSE electricity consumption Combination of LSTM with autoencoders”