Graph-enabled Reinforcement Learning for Time Series Forecasting with Adaptive Intelligence

TL;DR

This paper introduces a novel GraphRL framework combining Graph Neural Networks and Reinforcement Learning for improved time-series forecasting, especially in complex temporal structures like healthcare and weather data.

Contribution

The study presents a new GraphRL approach that integrates GNNs with RL for time-series prediction and demonstrates its superiority over traditional models.

Findings

GraphRL outperforms baseline models in forecasting accuracy.

GNNs effectively capture temporal dependencies in complex data.

Bayesian optimization further enhances model performance.

Abstract

Reinforcement learning is well known for its ability to model sequential tasks and learn latent data patterns adaptively. Deep learning models have been widely explored and adopted in regression and classification tasks. However, deep learning has its limitations such as the assumption of equally spaced and ordered data, and the lack of ability to incorporate graph structure in terms of time-series prediction. Graphical neural network (GNN) has the ability to overcome these challenges and capture the temporal dependencies in time-series data. In this study, we propose a novel approach for predicting time-series data using GNN and monitoring with Reinforcement Learning (RL). GNNs are able to explicitly incorporate the graph structure of the data into the model, allowing them to capture temporal dependencies in a more natural way. This approach allows for more accurate predictions in complex temporal structures, such as those found in healthcare, traffic and weather forecasting. We also fine-tune our GraphRL model using a Bayesian optimisation technique to further improve performance. The proposed framework outperforms the baseline models in time-series forecasting and monitoring. The contributions of this study include the introduction of a novel GraphRL framework for time-series prediction and the demonstration of the effectiveness of GNNs in comparison to traditional deep learning models such as RNNs and LSTMs. Overall, this study demonstrates the potential of GraphRL in providing accurate and efficient predictions in dynamic RL environments.