Conceitos essenciais
Enhancing QTS with AE-QTS for improved search performance in solving knapsack problems.
Resumo
In this paper, an enhanced version of the Quantum-inspired Tabu Search (QTS) algorithm is introduced as the "amplitude-ensemble" QTS (AE-QTS). This modification aims to improve the utilization of population information, resembling the Grover Search Algorithm while maintaining algorithm simplicity. The AE-QTS is applied to solve the classical combinatorial optimization 0/1 knapsack problem. Experimental results demonstrate that AE-QTS outperforms other algorithms, including QTS, by an average of 20% and up to 30% in some cases. Even with increasing problem complexity, AE-QTS consistently provides superior solutions compared to QTS. The study highlights that AE-QTS exhibits better search performance and efficiency.
Metaheuristics have been a popular research area due to their ability to solve complex problems like NP-complete and NP-hard problems. Quantum Algorithms have also shown significant advancements, leading researchers to integrate quantum characteristics into metaheuristic algorithms. Among these quantum-inspired algorithms, QEA and QTS are mainly conceived from a quantum perspective. While QEA leans towards traditional population-based thinking, QTS mirrors the Grover Search Algorithm conceptually.
The basic unit of a quantum computer is a qubit, which can exist in superposition states representing infinite information possibilities. Quantum logic gates perform reversible operations using rotation matrices for unitary operations. The 0/1 knapsack problem involves selecting items within weight constraints to maximize profit. Three cases categorize different methods of generating weights and profits for items.
The AE-QTS method expands on QTS by incorporating all explored information into qubits' amplitude adjustments across the entire population of each iteration. Experiments comparing AE-QTS with classical algorithms like GA, TS, DE show superior convergence results with increased efficiency over time. Performance improvements are notable across different item quantities and complexities.
The study concludes that incorporating population concepts from the Grover Search Algorithm into all qubits of AE-QTS enhances search performance compared to QTS. The proposed method maintains simplicity without adding parameters or complexity to existing implementations based on QTS.
Estatísticas
Experimental results show that AE-QTS outperforms other algorithms by at least an average of 20% in all cases and even by 30% in some cases.
Citações
"AE-QTS demonstrates better search performance and efficiency compared to other algorithms."
"Results prove that our method has better search performance."
"The 'amplitude-ensemble' mechanism increases the performance of QTS by approximately 20-30%."