A Hybrid Bayesian-Genetic Algorithm Based Hyperparameter Optimization of a LSTM Network for Demand Forecasting of Retail Products

Pravin Suryawanshi, Sandesh Gaikwad, Akansha Kumar, Akhil Patlolla, Sai Jayakumar

2023

Abstract

Demand forecasting is highly influenced by the non-linearity of time series data. Deep neural networks such as long short-term memory networks (LSTM) are considered better forecasters of such data. However, the LSTM network’s performances are subject to hyperparameter values. This study proposes a hybrid approach to determine the optimal set of hyperparameters of an LSTM model using Bayesian optimization and genetic algorithm. Bayesian optimization explores the search space in the direction where the improvement over the existing solution is likely, based on a fitness function. At the same time, a genetic algorithm is an evolutionary approach that can achieve global convergence by using selection, crossover, and mutation operators. The proposed hybrid approach utilizes the strengths of both these algorithms to tune the values of the hyperparameter of the LSTM network to minimize the forecasting error. In the dataset considered, we found that the hybrid approach reduced the forecasting error by approximately 27% compared to the Bayesian optimization approach. Additionally, the proposed method is better than the genetic algorithm when performed independently, with a decrease in error value by approximately 13%.

Paper Citation

in Harvard Style

Suryawanshi P., Gaikwad S., Kumar A., Patlolla A. and Jayakumar S. (2023). A Hybrid Bayesian-Genetic Algorithm Based Hyperparameter Optimization of a LSTM Network for Demand Forecasting of Retail Products. In Proceedings of the 15th International Joint Conference on Computational Intelligence - Volume 1: ECTA; ISBN 978-989-758-674-3, SciTePress, pages 230-237. DOI: 10.5220/0012182900003595

in Bibtex Style

@conference{ecta23,
author={Pravin Suryawanshi and Sandesh Gaikwad and Akansha Kumar and Akhil Patlolla and Sai Jayakumar},
title={A Hybrid Bayesian-Genetic Algorithm Based Hyperparameter Optimization of a LSTM Network for Demand Forecasting of Retail Products},
booktitle={Proceedings of the 15th International Joint Conference on Computational Intelligence - Volume 1: ECTA},
year={2023},
pages={230-237},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012182900003595},
isbn={978-989-758-674-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Joint Conference on Computational Intelligence - Volume 1: ECTA
TI - A Hybrid Bayesian-Genetic Algorithm Based Hyperparameter Optimization of a LSTM Network for Demand Forecasting of Retail Products
SN - 978-989-758-674-3
AU - Suryawanshi P.
AU - Gaikwad S.
AU - Kumar A.
AU - Patlolla A.
AU - Jayakumar S.
PY - 2023
SP - 230
EP - 237
DO - 10.5220/0012182900003595
PB - SciTePress