Losing dimensions: Geometric memorization in generative diffusion

• URL: http://arxiv.org/abs/2410.08727v1
• Date: Fri, 11 Oct 2024 11:31:20 GMT
• Title: Losing dimensions: Geometric memorization in generative diffusion
• Authors: Beatrice Achilli, Enrico Ventura, Gianluigi Silvestri, Bao Pham, Gabriel Raya, Dmitry Krotov, Carlo Lucibello, Luca Ambrogioni,
• Abstract summary: We extend the theory of memorization in generative diffusion to manifold-supported data. Our theoretical and experimental findings indicate that different subspaces are lost due to memorization effects at different critical times and dataset sizes. Perhaps counterintuitively, we find that, under some conditions, subspaces of higher variance are lost first due to memorization effects.
• Score: 10.573546162574235
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Generative diffusion processes are state-of-the-art machine learning models deeply connected with fundamental concepts in statistical physics. Depending on the dataset size and the capacity of the network, their behavior is known to transition from an associative memory regime to a generalization phase in a phenomenon that has been described as a glassy phase transition. Here, using statistical physics techniques, we extend the theory of memorization in generative diffusion to manifold-supported data. Our theoretical and experimental findings indicate that different tangent subspaces are lost due to memorization effects at different critical times and dataset sizes, which depend on the local variance of the data along their directions. Perhaps counterintuitively, we find that, under some conditions, subspaces of higher variance are lost first due to memorization effects. This leads to a selective loss of dimensionality where some prominent features of the data are memorized without a full collapse on any individual training point. We validate our theory with a comprehensive set of experiments on networks trained both in image datasets and on linear manifolds, which result in a remarkable qualitative agreement with the theoretical predictions.

Related papers

• Generalization Below the Edge of Stability: The Role of Data Geometry [60.147710896851045]
  We show how data geometry controls generalization in ReLU networks trained below the edge of stability.<n>For data distributions supported on a mixture of low-dimensional balls, we derive generalization bounds that provably adapt to the intrinsic dimension.<n>Our results consolidate disparate empirical findings that have appeared in the literature.
  arXiv  Detail & Related papers  (2025-10-20T21:40:36Z)
• On the Edge of Memorization in Diffusion Models [25.927892368310868]
  We introduce a scientific and mathematical "laboratory" for investigating memorization and generalization in practical diffusion models.<n>Our work provides an analytically tractable and practically meaningful setting for future theoretical and empirical investigations.
  arXiv  Detail & Related papers  (2025-08-25T05:56:05Z)
• Bigger Isn't Always Memorizing: Early Stopping Overparameterized Diffusion Models [51.03144354630136]
  Generalization in natural data domains is progressively achieved during training before the onset of memorization.<n>Generalization vs. memorization is then best understood as a competition between time scales.<n>We show that this phenomenology is recovered in diffusion models learning a simple probabilistic context-free grammar with random rules.
  arXiv  Detail & Related papers  (2025-05-22T17:40:08Z)
• Cross-Entropy Is All You Need To Invert the Data Generating Process [29.94396019742267]
  Empirical phenomena suggest that supervised models can learn interpretable factors of variation in a linear fashion. Recent advances in self-supervised learning have shown that these methods can recover latent structures by inverting the data generating process. We prove that even in standard classification tasks, models learn representations of ground-truth factors of variation up to a linear transformation.
  arXiv  Detail & Related papers  (2024-10-29T09:03:57Z)
• Learning Discrete Concepts in Latent Hierarchical Models [73.01229236386148]
  Learning concepts from natural high-dimensional data holds potential in building human-aligned and interpretable machine learning models. We formalize concepts as discrete latent causal variables that are related via a hierarchical causal model. We substantiate our theoretical claims with synthetic data experiments.
  arXiv  Detail & Related papers  (2024-06-01T18:01:03Z)
• Minimum-Norm Interpolation Under Covariate Shift [14.863831433459902]
  In-distribution research on high-dimensional linear regression has led to the identification of a phenomenon known as textitbenign overfitting We prove the first non-asymptotic excess risk bounds for benignly-overfit linear interpolators in the transfer learning setting.
  arXiv  Detail & Related papers  (2024-03-31T01:41:57Z)
• Dynamical Regimes of Diffusion Models [14.797301819675454]
  We study generative diffusion models in the regime where the dimension of space and the number of data are large. Our analysis reveals three distinct dynamical regimes during the backward generative diffusion process. The dependence of the collapse time on the dimension and number of data provides a thorough characterization of the curse of dimensionality for diffusion models.
  arXiv  Detail & Related papers  (2024-02-28T17:19:26Z)
• Towards Theoretical Understandings of Self-Consuming Generative Models [56.84592466204185]
  This paper tackles the emerging challenge of training generative models within a self-consuming loop. We construct a theoretical framework to rigorously evaluate how this training procedure impacts the data distributions learned by future models. We present results for kernel density estimation, delivering nuanced insights such as the impact of mixed data training on error propagation.
  arXiv  Detail & Related papers  (2024-02-19T02:08:09Z)
• Assessing Neural Network Representations During Training Using Noise-Resilient Diffusion Spectral Entropy [55.014926694758195]
  Entropy and mutual information in neural networks provide rich information on the learning process. We leverage data geometry to access the underlying manifold and reliably compute these information-theoretic measures. We show that they form noise-resistant measures of intrinsic dimensionality and relationship strength in high-dimensional simulated data.
  arXiv  Detail & Related papers  (2023-12-04T01:32:42Z)
• On Memorization in Diffusion Models [46.656797890144105]
  We show that memorization behaviors tend to occur on smaller-sized datasets. We quantify the impact of the influential factors on these memorization behaviors in terms of effective model memorization (EMM) Our study holds practical significance for diffusion model users and offers clues to theoretical research in deep generative models.
  arXiv  Detail & Related papers  (2023-10-04T09:04:20Z)
• Towards a mathematical understanding of learning from few examples with nonlinear feature maps [68.8204255655161]
  We consider the problem of data classification where the training set consists of just a few data points. We reveal key relationships between the geometry of an AI model's feature space, the structure of the underlying data distributions, and the model's generalisation capabilities.
  arXiv  Detail & Related papers  (2022-11-07T14:52:58Z)
• Learning from few examples with nonlinear feature maps [68.8204255655161]
  We explore the phenomenon and reveal key relationships between dimensionality of AI model's feature space, non-degeneracy of data distributions, and the model's generalisation capabilities. The main thrust of our present analysis is on the influence of nonlinear feature transformations mapping original data into higher- and possibly infinite-dimensional spaces on the resulting model's generalisation capabilities.
  arXiv  Detail & Related papers  (2022-03-31T10:36:50Z)
• Multi-scale Feature Learning Dynamics: Insights for Double Descent [71.91871020059857]
  We study the phenomenon of "double descent" of the generalization error. We find that double descent can be attributed to distinct features being learned at different scales.
  arXiv  Detail & Related papers  (2021-12-06T18:17:08Z)
• On the geometry of generalization and memorization in deep neural networks [15.250162344382051]
  We study the structure of when and where memorization occurs in a deep network. All layers preferentially learn from examples which share features, and link this behavior to generalization performance. We find that memorization predominately occurs in the deeper layers, due to decreasing object' radius and dimension.
  arXiv  Detail & Related papers  (2021-05-30T19:07:33Z)
• More data or more parameters? Investigating the effect of data structure on generalization [17.249712222764085]
  Properties of data impact the test error as a function of the number of training examples and number of training parameters. We show that noise in the labels and strong anisotropy of the input data play similar roles on the test error.
  arXiv  Detail & Related papers  (2021-03-09T16:08:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.