Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

Booming noise, along with other factors contributing to interior vehicle noise, significantly influences the overall perception of an automobile's quality. In light of the automotive industry's strong focus on customer satisfaction, developing robust sound quality prediction models and comprehensive testing procedures is of paramount importance. These endeavors assist manufacturers in delivering a more enjoyable and comfortable driving experience, ultimately strengthening their vehicles' reputation in the market. However, the datasets used for training sound quality prediction models often exhibit class imbalances, posing a substantial challenge for machine learning. This imbalance can result in diminished predictive performance, especially for the minority class. Booming noise detection is particularly sensitive to this issue, given the greater prevalence of normal conditions compared to faulty ones. In our study, we address imbalanced learning across three vehicle types and investigate various imbalance ratios. Our findings highlight the effectiveness of a novel reinforcement learning framework in enhancing model generalizability and robustness to noise in the context of imbalanced data, outperforming traditional methods in these aspects.

Original publication




Conference paper

Publication Date



24 - 29