Identifying Memorization of Diffusion Models through p-Laplace Analysis
This addresses the problem of detecting memorization in generative models for researchers and practitioners concerned with model privacy and data leakage.
This work tackles the problem of identifying memorized training data in diffusion models by leveraging estimated score functions to compute p-Laplace operators, demonstrating effectiveness in both Gaussian mixture models and image generative models.
Diffusion models, today's leading image generative models, estimate the score function, i.e. the gradient of the log probability of (perturbed) data samples, without direct access to the underlying probability distribution. This work investigates whether the estimated score function can be leveraged to compute higher-order differentials, namely p-Laplace operators. We show here these operators can be employed to identify memorized training data. We propose a numerical p-Laplace approximation based on the learned score functions, showing its effectiveness in identifying key features of the probability landscape. We analyze the structured case of Gaussian mixture models, and demonstrate the results carry-over to image generative models, where memorization identification based on the p-Laplace operator is performed for the first time.