Feature Space Perturbation: A Panacea to Enhanced Transferability Estimation
This work addresses the challenge of model selection for transfer learning, offering a more robust estimation method, though it appears incremental as it builds on prior metrics like LogMe.
The paper tackles the problem of selecting optimal pre-trained models for downstream tasks by introducing a feature perturbation method that enhances transferability estimation, resulting in a 28.84% performance improvement over the existing LogMe method.
Leveraging a transferability estimation metric facilitates the non-trivial challenge of selecting the optimal model for the downstream task from a pool of pre-trained models. Most existing metrics primarily focus on identifying the statistical relationship between feature embeddings and the corresponding labels within the target dataset, but overlook crucial aspect of model robustness. This oversight may limit their effectiveness in accurately ranking pre-trained models. To address this limitation, we introduce a feature perturbation method that enhances the transferability estimation process by systematically altering the feature space. Our method includes a Spread operation that increases intra-class variability, adding complexity within classes, and an Attract operation that minimizes the distances between different classes, thereby blurring the class boundaries. Through extensive experimentation, we demonstrate the efficacy of our feature perturbation method in providing a more precise and robust estimation of model transferability. Notably, the existing LogMe method exhibited a significant improvement, showing a 28.84% increase in performance after applying our feature perturbation method.