Time series forecasting using a deep belief network with restricted Boltzmann machines

By Takashi Kuremoto, Shinsuke Kimura, Kunikazu Kobayashi, Masanao Obayashi. 2014

511 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering, Management Science and Operations Research

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. Henry Leung
2. Jun Zhao
3. King Fai Ma
4. Zhongyang Han
5. Wei Wang
6. Daniel Hsu
7. Jian Lian
8. A. K. Qin

In Electrical and Electronic Engineering

1. Daniel Hsu
2. Jian Lian
3. Henry Leung
4. Dong-Sik Jang
5. Michael Fairbank
6. Sangsung Park
7. Ljupco Kocarev
8. Weikuan Jia

In Management Science and Operations Research

1. Daniel Hsu
2. Henry Leung
3. Dong-Sik Jang
4. Michael Fairbank
5. Kay Chen Tan
6. Sangsung Park
7. Sunita Sarawagi
8. Ljupco Kocarev

Citations in the corpus

• Research on exchange rate forecasting based on deep belief network (2017, unknown): “[7] studied the applications of deep learning on time series problem and the use of particle swarm optimization to optimize the number of hidden layers and nodes in DBN.”
• Feature dimensionality reduction: a review (2022, methodology): “To optimize the initialization and local optimization in time series prediction, a deep belief network was applied to predict time series, which was applied to the approximation and short-term prediction of chaotic time series such as a logistic map and Lorenz chaos [161].”
• Deep Learning-Based Corporate Performance Prediction Model Considering Technical Capability (2017, background): “Kuremoto et al. (2014) proposed a modified DBN-based model for time series forecasting [39].”
• ARMDN: Associative and Recurrent Mixture Density Networks for eRetail Demand Forecasting (2018, methodology): “Takashi Kuremoto used multiple Restricted Boltzmann machines (RBM) optimized via Particle Swarm Optimization (PSO) [18].”
• Stock prediction using deep learning (2016, methodology): “DNNs have been applied to some time series forecasting and they have shown good results [11, 12].”
• Multi-period Time Series Modeling with Sparsity via Bayesian Variational Inference (2017, methodology): “However, as compared in [12], deep learning models such as RBM and SAE perform worse than LSTM because they cannot capture long-term dependencies across data points, and fixedsize input window also leads to sub-optimal performance.”
• Deep Learning for Time-Series Analysis (2017, methodology): “For example, Deep Belief Networks are used in the work of [33] along with RBM. [58] also compares the performance of Deep Belief Networks with that of Stacked Denoising Autoencoders.”
• Convolutional neural networks applied to high-frequency market microstructure forecasting (2017, methodology): “[10] used Restricted Boltzmann Machines and achieved decent results in modelling chaotic time series.”