Reconstructing Training Data From Real World Models Trained with Transfer Learning
This work addresses a potential privacy risk for data leakage in real-world models trained with transfer learning, representing a novel method for a known bottleneck.
The paper tackles the problem of reconstructing training data from classifiers in realistic settings, enabling data reconstruction for models trained on high-resolution images using transfer learning, with a novel clustering method that improves identification of good reconstructions without prior knowledge of the training set.
Current methods for reconstructing training data from trained classifiers are restricted to very small models, limited training set sizes, and low-resolution images. Such restrictions hinder their applicability to real-world scenarios. In this paper, we present a novel approach enabling data reconstruction in realistic settings for models trained on high-resolution images. Our method adapts the reconstruction scheme of arXiv:2206.07758 to real-world scenarios -- specifically, targeting models trained via transfer learning over image embeddings of large pre-trained models like DINO-ViT and CLIP. Our work employs data reconstruction in the embedding space rather than in the image space, showcasing its applicability beyond visual data. Moreover, we introduce a novel clustering-based method to identify good reconstructions from thousands of candidates. This significantly improves on previous works that relied on knowledge of the training set to identify good reconstructed images. Our findings shed light on a potential privacy risk for data leakage from models trained using transfer learning.