Bigger Isn't Always Memorizing: Early Stopping Overparameterized Diffusion Models

• URL: http://arxiv.org/abs/2505.16959v1
• Date: Thu, 22 May 2025 17:40:08 GMT
• Title: Bigger Isn't Always Memorizing: Early Stopping Overparameterized Diffusion Models
• Authors: Alessandro Favero, Antonio Sclocchi, Matthieu Wyart,
• Abstract summary: 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.
• Score: 51.03144354630136
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Diffusion probabilistic models have become a cornerstone of modern generative AI, yet the mechanisms underlying their generalization remain poorly understood. In fact, if these models were perfectly minimizing their training loss, they would just generate data belonging to their training set, i.e., memorize, as empirically found in the overparameterized regime. We revisit this view by showing that, in highly overparameterized diffusion models, generalization in natural data domains is progressively achieved during training before the onset of memorization. Our results, ranging from image to language diffusion models, systematically support the empirical law that memorization time is proportional to the dataset size. Generalization vs. memorization is then best understood as a competition between time scales. We show that this phenomenology is recovered in diffusion models learning a simple probabilistic context-free grammar with random rules, where generalization corresponds to the hierarchical acquisition of deeper grammar rules as training time grows, and the generalization cost of early stopping can be characterized. We summarize these results in a phase diagram. Overall, our results support that a principled early-stopping criterion - scaling with dataset size - can effectively optimize generalization while avoiding memorization, with direct implications for hyperparameter transfer and privacy-sensitive applications.

Related papers

• Generalization Capability for Imitation Learning [1.30536490219656]
  Imitation learning holds the promise of equipping robots with versatile skills by learning from expert demonstrations.<n>However, policies trained on finite datasets often struggle to generalize beyond the training distribution.<n>We present a unified perspective on the generalization capability of imitation learning, grounded in both information theorey and data distribution property.
  arXiv  Detail & Related papers  (2025-04-25T17:59:59Z)
• AutoElicit: Using Large Language Models for Expert Prior Elicitation in Predictive Modelling [53.54623137152208]
  We introduce AutoElicit to extract knowledge from large language models and construct priors for predictive models.<n>We show these priors are informative and can be refined using natural language.<n>We find that AutoElicit yields priors that can substantially reduce error over uninformative priors, using fewer labels, and consistently outperform in-context learning.
  arXiv  Detail & Related papers  (2024-11-26T10:13:39Z)
• Causal Estimation of Memorisation Profiles [58.20086589761273]
  Understanding memorisation in language models has practical and societal implications. Memorisation is the causal effect of training with an instance on the model's ability to predict that instance. This paper proposes a new, principled, and efficient method to estimate memorisation based on the difference-in-differences design from econometrics.
  arXiv  Detail & Related papers  (2024-06-06T17:59:09Z)
• Skews in the Phenomenon Space Hinder Generalization in Text-to-Image Generation [59.138470433237615]
  We introduce statistical metrics that quantify both the linguistic and visual skew of a dataset for relational learning. We show that systematically controlled metrics are strongly predictive of generalization performance. This work informs an important direction towards quality-enhancing the data diversity or balance to scaling up the absolute size.
  arXiv  Detail & Related papers  (2024-03-25T03:18:39Z)
• A Temporally Disentangled Contrastive Diffusion Model for Spatiotemporal Imputation [35.46631415365955]
  We introduce a conditional diffusion framework called C$2$TSD, which incorporates disentangled temporal (trend and seasonality) representations as conditional information. Our experiments on three real-world datasets demonstrate the superior performance of our approach compared to a number of state-of-the-art baselines.
  arXiv  Detail & Related papers  (2024-02-18T11:59:04Z)
• Non-Vacuous Generalization Bounds for Large Language Models [78.42762571499061]
  We provide the first non-vacuous generalization bounds for pretrained large language models. We show that larger models have better generalization bounds and are more compressible than smaller models.
  arXiv  Detail & Related papers  (2023-12-28T17:58:42Z)
• On Memorization in Diffusion Models [44.031805633114985]
  We show that memorization behaviors tend to occur on smaller-sized datasets.<n>We quantify the impact of the influential factors on these memorization behaviors in terms of effective model memorization (EMM)<n>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)
• On the Generalization of Diffusion Model [42.447639515467934]
  We define the generalization of the generative model, which is measured by the mutual information between the generated data and the training set. We show that for the empirical optimal diffusion model, the data generated by a deterministic sampler are all highly related to the training set, thus poor generalization. We propose another training objective whose empirical optimal solution has no potential generalization problem.
  arXiv  Detail & Related papers  (2023-05-24T04:27:57Z)
• Generalization and Memorization: The Bias Potential Model [9.975163460952045]
  generative models and density estimators behave quite differently from models for learning functions. For the bias potential model, we show that dimension-independent generalization accuracy is achievable if early stopping is adopted. In the long term, the model either memorizes the samples or diverges.
  arXiv  Detail & Related papers  (2020-11-29T04:04:54Z)
• Whitening and second order optimization both make information in the dataset unusable during training, and can reduce or prevent generalization [50.53690793828442]
  We show that both data whitening and second order optimization can harm or entirely prevent generalization. For a general class of models, namely models with a fully connected first layer, we prove that the information contained in this matrix is the only information which can be used to generalize.
  arXiv  Detail & Related papers  (2020-08-17T18:00:05Z)

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.