2025 IEEE International Conference on Cyber Security and Resilience

Full Program

Summary:

Traditional fuzzing techniques generate these inputs by randomly mutating predefined seed inputs. This random mutation technique operates without knowledge of the target program making it inefficient in achieving high code coverage and detecting vulnerabilities. Reinforcement learning (RL) has emerged as a promising approach to address these limitations by learning the optimal mutation strategies. However, existing reinforcement learning-based fuzzing methods do not handle the mutation of multi-parametric test cases where multiple parameters require simultaneous yet distinct mutation strategies. This paper presents a reinforcement learning-based fuzzing technique that optimizes the mutation strategy for multi-parametric test cases.
By dynamically adapting mutation strategies the proposed approach improves the effectiveness of the fuzzing process increases the identification of bugs and enhances code coverage.
We evaluate the proposed method against traditional fuzzing techniques and an existing reinforcement learning-based fuzzing approach, demonstrating significant improvements in coverage metrics and crash detection rates.

Author(s):

Marie Louise Uwibambe    
United States

Akanksha Tyagi    
University of Arkansas
United States

Qinghua Li    
University of Arkansas
United States

 


Copyright © 2025 SUMMIT-TEC GROUP LTD