Uso de computación cuántica en la mejora de algoritmos de aprendizaje automático
Paulo César Galarza-Sánchez, Alex Fernando Erazo-Luzuriaga, Miguel Fabricio Boné-Andrade · Revista Científica Ciencia y Método · 2023
This study explores the integration of quantum computing in the improvement of machine learning algorithms, highlighting its potential to overcome the computational limitations of classical methods in highly complex tasks. Through a systematic literature review with a qualitative approach, academic articles indexed between 2015 and 2023 were analyzed, identifying key advances in variational quantum algorithms, quantum support vector machines, and quantum neural networks. The findings reveal that, despite current technological constraints, these approaches show advantages in efficiency, representativeness and generalizability. Furthermore, the relevance of hybrid quantum-classical models as an intermediate solution is highlighted, by allowing a functional distribution between quantum and classical resources. The research concludes that this convergence represents a promising avenue for the development of advanced artificial intelligence, although challenges such as circuit optimization, noise mitigation and methodological standardization remain. The need to strengthen the technological and theoretical infrastructure to consolidate this line of computational innovation is emphasized.