Skip to content

Latest commit

 

History

History
13 lines (7 loc) · 2.31 KB

README.md

File metadata and controls

13 lines (7 loc) · 2.31 KB

Using a Quantum Genetic Algorithm to Optimize the Spending of a Cyber Security Budget

Abstract:

As difficult choices amidst technological advance- ments highlight the importance of optimization, the single-choice Knapsack problem emerges as a valuable model for scenarios where inputs are discrete and limited. In this paper, I propose a novel algorithm, Quantum Save, that uses quantum computing and genetic algorithms to optimize this Knapsack problem. With IBM’s Qiskit Toolkit, Quantum Save begins by initializing a qubit population and using distributions from quantum measurement as inputs for the Knapsack problem. Reproduction is then sim- ulated by adjusting probability amplitudes based on Knapsack output and implementing a disaster algorithm. After comprehen- sive testing of Quantum Save with different hyper-parameters, a probability amplitude reduction of 10% and a disaster-step of 5 generations was chosen. Quantum Save’s rearrangement of qubits and hyper-parameter tuning resulted in a 49.2% improvement of Knapsack output compared to current implementations of quantum-genetic algorithms. Qubit representation of discrete in- puts helps store multiple states while maintaining high accuracy. To emphasize Quantum Save’s efficacy in real-world applications, this paper analyzes the algorithm’s performance in allocating lim- ited budgets. Using patterns in data collected from recent cyber attacks, Quantum Save suggested the implementation of specific budget-friendly controls to minimize a company’s damages. By analyzing the distribution of countermeasures selected at each budget range, there was strong evidence that Quantum Save increased mitigation from attacks. Also, Quantum Save scales linearly, with O(n) complexity, making it fast and efficient. So, in the context of this case-study, Quantum Save harnesses the computational advantage of quantum computing to effectively optimize the Knapsack problem.

Code Structure:

In this repository, all code is placed in the "Code Folder" and the background research inspiring the work is placed in the "Background Research Articles". Within the "Code Folder", there are various steps of the project. My main paper is "Hariharan QCE Paper 20". Other folders are adequately named.

Contact Information:

If you have any questions, feel free to reach out at -- [email protected]