دانشگاه آزاد اسلامی واحد تهران جنوب

صفحه اصلی / نخستین همایش هوش مصنوعی و پژوهش های نوظهور: همگرایی انسان و سیستم های هوشمند

Quantum-Inspired Evolutionary Algorithm for Hyperparameter Tuning in Automatic Number Plate Recognition Systems

نویسندگان :

Mazyar Taghavi¹ Amir Mohammad Paslari² Mahdi Riazi³

1- Iran University of Science and Technology 2- Iran University of Science and Technology 3- K. N. Toosi University of Technology

کلمات کلیدی :

Automatic Number Plate Recognition،Quantum-Inspired Optimization،Hyperparameter Tuning

چکیده :

Hyperparameter tuning is a critical step in optimizing machine learning models, particularly for complex tasks such as Automatic Number Plate Recognition (ANPR). Traditional methods like grid search and random search are computationally expensive and inefficient for high-dimensional search spaces. This paper proposes a novel approach using a Quantum-Inspired Evolutionary Algorithm (QEA) to optimize hyperparameters for an ANPR system designed for Persian language numbers. The QEA leverages quantum-inspired mechanisms such as superposition and entanglement to explore the hyperparameter space more efficiently. We evaluate the QEA on a standard dataset and compare its performance against a classical evolutionary algorithm. Results demonstrate that the QEA achieves higher model accuracy, faster convergence, and better computational efficiency, making it a promising tool for hyperparameter optimization in machine learning. Codes are available on the GitHub repository.