Quantum Neural Network model for Token allocation for Course Bidding

Quantum computer has shown the advantage over the classical computer to solve some problems using the laws of quantum mechanics. With a combination of knowledge of machine learning and quantum computing, Quantum neural networks adapted the concept from classical neural networks and apply parameterized quantum gates as neural network weights. In this paper, we present an application of quantum neural networks with real-world data to predict token price used in a course bidding system. The experiments were carried out on the Qiskit quantum simulator. The result shows that quantum neural networks can achieve a good prediction result compared to the classical neural network. The best model configuration has the lowest RMSE 6.38%. This approach opens an opportunity to explore the benefit of quantum machine learning in many research fields in the future.