Learning from multivariate discrete sequential data using a restricted Boltzmann machine model

TL;DR

This paper introduces the p-RBM, a memory-capable extension of the RBM, designed for modeling multivariate discrete sequential data, demonstrated through stock market prediction with promising results.

Contribution

The paper proposes the p-RBM, a novel extension of RBM that retains memory of past states, enabling effective modeling of dynamic sequential data.

Findings

p-RBM outperforms traditional RBM in sequence prediction tasks

The model successfully predicts stock market directions

Promising results in modeling multivariate discrete time-series

Abstract

A restricted Boltzmann machine (RBM) is a generative neural-network model with many novel applications such as collaborative filtering and acoustic modeling. An RBM lacks the capacity to retain memory, making it inappropriate for dynamic data modeling as in time-series analysis. In this paper we address this issue by proposing the p-RBM model, a generalization of the regular RBM model, capable of retaining memory of p past states. We further show how to train the p-RBM model using contrastive divergence and test our model on the problem of predicting the stock market direction considering 100 stocks of the NASDAQ-100 index. Obtained results show that the p-RBM offer promising prediction potential.