Image Shortcut Squeezing: Countering Perturbative Availability Poisons with Compression
This addresses the challenge of ensuring robust model training against poisoning attacks for machine learning practitioners, though it appears incremental as it builds on existing countermeasure approaches.
The paper tackles the problem of countering perturbative availability poisons (PAPs) that hinder image-based model training, showing that Image Shortcut Squeezing (ISS) using compression restores CIFAR-10 accuracy to 81.73%, outperforming previous methods by 37.97%.
Perturbative availability poisons (PAPs) add small changes to images to prevent their use for model training. Current research adopts the belief that practical and effective approaches to countering PAPs do not exist. In this paper, we argue that it is time to abandon this belief. We present extensive experiments showing that 12 state-of-the-art PAP methods are vulnerable to Image Shortcut Squeezing (ISS), which is based on simple compression. For example, on average, ISS restores the CIFAR-10 model accuracy to $81.73\%$, surpassing the previous best preprocessing-based countermeasures by $37.97\%$ absolute. ISS also (slightly) outperforms adversarial training and has higher generalizability to unseen perturbation norms and also higher efficiency. Our investigation reveals that the property of PAP perturbations depends on the type of surrogate model used for poison generation, and it explains why a specific ISS compression yields the best performance for a specific type of PAP perturbation. We further test stronger, adaptive poisoning, and show it falls short of being an ideal defense against ISS. Overall, our results demonstrate the importance of considering various (simple) countermeasures to ensure the meaningfulness of analysis carried out during the development of PAP methods.