Vehicle interior noise prediction improved with cheaper labeled data
SpecRegMatch: Robust Semi-Supervised Regression for Vehicle Interior Noise Prediction
Artificial IntelligenceComputer Vision and Pattern Recognition
Summary
Measuring inside car noise can help make rides quieter, but collecting enough labeled data for accurate predictions is expensive. The authors created SpecRegMatch, a method that predicts noise levels using both a small amount of labeled data and plenty of unlabeled data, without needing to train many separate models. Their approach uses clever tricks to keep predictions stable even when the input data is changed slightly. This method helps improve prediction accuracy, especially when labeled data is limited.
What this means in practice
- •For automotive engineers: Improve vehicle cabin noise prediction models when labeled acoustic data is scarce, reducing testing cost.
- •For machine learning engineers: Reduce computational load when training semi-supervised regression models by using a single model approach with data augmentations.
Authors
Sejin Sim, Jinsoo Bae, Seoung Bum Kim
Abstract
The rapid advancement of artificial intelligence has observed increased application in predicting vehicle interior noise levels within the automotive industry. However, the collection of labeled data for training models in this context involves significant costs. Previous studies in semi-supervised regression (SSR) have effectively mitigated the reliance on labeled data by incorporating unlabeled data. Nonetheless, these approaches often introduce a high computational cost due to the training of multiple models and data sampling. This study introduces SpecRegMatch, a novel SSR method aimed at addressing the computational cost associated with training by leveraging a single model, thus eliminating the need for multiple data samplings. SpecRegMatch integrates consistency regularization and information maximization to robustly train the model, achieved through various augmentations applied to both the embedding vectors and predicted values. Experimental results demonstrate that SpecRegMatch achieves state-of-the-art performance across various scenarios, even when using a single model. It attains a remarkable performance, as indicated by an R^2 score of 0.434. This is especially noteworthy in scenarios where labeled data is scarce. You can access the code for our proposed method at https://github.com/sejin-sim/SpecRegMatch.