A Hybrid Quantum Deep Learning Approach Based on Intelligent Optimization to Predict the Broiler Energies

The global population has undergone rapid growth, particularly in the second half of the last century; consequently, total meat production is increasing rapidly. Deep neural networks have been used to solve many problems in different areas. The parameters of a deep learning model have a significant impact on the ability of the model to map relationships between the input and output data. Thus, many techniques are used to determine and optimise these parameters. Quantum computing is a rapidly growing discipline that attracts the attention of large number of researchers. Although classical computers have limitations, quantum computing helps to overcome these limitations and promises a step-change in computational performance. This paper proposes the combination of the quantum deep learning model (QDL) with the genetic algorithm technique (GA) to determine the best values of the parameters in QDL networks. Data were collected for this study from 210 broiler farms in Mazandaran, Iran. The results show that the $$R^2$$ R 2 indexes for the QDL model at the test stage for broiler meat and manure outputs are 0.81 and 0.79, respectively.