What Secrets Do Your Manifolds Hold? Understanding the Local Geometry of Generative Models

• URL: http://arxiv.org/abs/2408.08307v2
• Date: Thu, 06 Feb 2025 09:30:06 GMT
• Title: What Secrets Do Your Manifolds Hold? Understanding the Local Geometry of Generative Models
• Authors: Ahmed Imtiaz Humayun, Ibtihel Amara, Cristina Vasconcelos, Deepak Ramachandran, Candice Schumann, Junfeng He, Katherine Heller, Golnoosh Farnadi, Negar Rostamzadeh, Mohammad Havaei,
• Abstract summary: We study the local geometry of the learned manifold and its relationship to generation outcomes for a wide range of generative models.<n>We provide quantitative and qualitative evidence showing that for a given latent-image pair, the local descriptors are indicative of generation aesthetics, diversity, and memorization by the generative model.
• Score: 17.273596999339077
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Deep Generative Models are frequently used to learn continuous representations of complex data distributions using a finite number of samples. For any generative model, including pre-trained foundation models with Diffusion or Transformer architectures, generation performance can significantly vary across the learned data manifold. In this paper we study the local geometry of the learned manifold and its relationship to generation outcomes for a wide range of generative models, including DDPM, Diffusion Transformer (DiT), and Stable Diffusion 1.4. Building on the theory of continuous piecewise-linear (CPWL) generators, we characterize the local geometry in terms of three geometric descriptors - scaling ($\psi$), rank ($\nu$), and complexity/un-smoothness ($\delta$). We provide quantitative and qualitative evidence showing that for a given latent-image pair, the local descriptors are indicative of generation aesthetics, diversity, and memorization by the generative model. Finally, we demonstrate that by training a reward model on the local scaling for Stable Diffusion, we can self-improve both generation aesthetics and diversity using `geometry reward' based guidance during denoising.

Related papers

• On the Statistical Capacity of Deep Generative Models [10.288413514555861]
  We show that deep generative models can only generate concentrated samples that exhibit light tails. These results shed light on the limited capacity of common deep generative models to handle heavy tails.
  arXiv  Detail & Related papers  (2025-01-14T00:39:46Z)
• SpaceMesh: A Continuous Representation for Learning Manifold Surface Meshes [61.110517195874074]
  We present a scheme to directly generate manifold, polygonal meshes of complex connectivity as the output of a neural network. Our key innovation is to define a continuous latent connectivity space at each mesh, which implies the discrete mesh. In applications, this approach not only yields high-quality outputs from generative models, but also enables directly learning challenging geometry processing tasks such as mesh repair.
  arXiv  Detail & Related papers  (2024-09-30T17:59:03Z)
• Latent diffusion models for parameterization and data assimilation of facies-based geomodels [0.0]
  Diffusion models are trained to generate new geological realizations from input fields characterized by random noise. Latent diffusion models are shown to provide realizations that are visually consistent with samples from geomodeling software.
  arXiv  Detail & Related papers  (2024-06-21T01:32:03Z)
• Heat Death of Generative Models in Closed-Loop Learning [63.83608300361159]
  We study the learning dynamics of generative models that are fed back their own produced content in addition to their original training dataset. We show that, unless a sufficient amount of external data is introduced at each iteration, any non-trivial temperature leads the model to degenerate.
  arXiv  Detail & Related papers  (2024-04-02T21:51:39Z)
• Compressing Image-to-Image Translation GANs Using Local Density Structures on Their Learned Manifold [69.33930972652594]
  Generative Adversarial Networks (GANs) have shown remarkable success in modeling complex data distributions for image-to-image translation. Existing GAN compression methods mainly rely on knowledge distillation or convolutional classifiers' pruning techniques. We propose a new approach by explicitly encouraging the pruned model to preserve the density structure of the original parameter-heavy model on its learned manifold. Our experiments on image translation GAN models, Pix2Pix and CycleGAN, with various benchmark datasets and architectures demonstrate our method's effectiveness.
  arXiv  Detail & Related papers  (2023-12-22T15:43:12Z)
• Improving Out-of-Distribution Robustness of Classifiers via Generative Interpolation [56.620403243640396]
  Deep neural networks achieve superior performance for learning from independent and identically distributed (i.i.d.) data. However, their performance deteriorates significantly when handling out-of-distribution (OoD) data. We develop a simple yet effective method called Generative Interpolation to fuse generative models trained from multiple domains for synthesizing diverse OoD samples.
  arXiv  Detail & Related papers  (2023-07-23T03:53:53Z)
• Geometric Neural Diffusion Processes [55.891428654434634]
  We extend the framework of diffusion models to incorporate a series of geometric priors in infinite-dimension modelling. We show that with these conditions, the generative functional model admits the same symmetry.
  arXiv  Detail & Related papers  (2023-07-11T16:51:38Z)
• T1: Scaling Diffusion Probabilistic Fields to High-Resolution on Unified Visual Modalities [69.16656086708291]
  Diffusion Probabilistic Field (DPF) models the distribution of continuous functions defined over metric spaces. We propose a new model comprising of a view-wise sampling algorithm to focus on local structure learning. The model can be scaled to generate high-resolution data while unifying multiple modalities.
  arXiv  Detail & Related papers  (2023-05-24T03:32:03Z)
• Geometric Latent Diffusion Models for 3D Molecule Generation [172.15028281732737]
  Generative models, especially diffusion models (DMs), have achieved promising results for generating feature-rich geometries. We propose a novel and principled method for 3D molecule generation named Geometric Latent Diffusion Models (GeoLDM)
  arXiv  Detail & Related papers  (2023-05-02T01:07:22Z)
• VTAE: Variational Transformer Autoencoder with Manifolds Learning [144.0546653941249]
  Deep generative models have demonstrated successful applications in learning non-linear data distributions through a number of latent variables. The nonlinearity of the generator implies that the latent space shows an unsatisfactory projection of the data space, which results in poor representation learning. We show that geodesics and accurate computation can substantially improve the performance of deep generative models.
  arXiv  Detail & Related papers  (2023-04-03T13:13:19Z)
• Conformal Generative Modeling on Triangulated Surfaces [31.870141076085716]
  We propose conformal generative modeling, a framework for generative modeling on 2D surfaces approximated by discrete triangle meshes. Our approach leverages advances in discrete conformal geometry to develop a map from a source triangle mesh to a target triangle mesh of a simple manifold such as a sphere.
  arXiv  Detail & Related papers  (2023-03-17T21:06:47Z)
• Semi-Supervised Manifold Learning with Complexity Decoupled Chart Autoencoders [45.29194877564103]
  This work introduces a chart autoencoder with an asymmetric encoding-decoding process that can incorporate additional semi-supervised information such as class labels. We discuss the approximation power of such networks and derive a bound that essentially depends on the intrinsic dimension of the data manifold rather than the dimension of ambient space.
  arXiv  Detail & Related papers  (2022-08-22T19:58:03Z)
• Unveiling the Latent Space Geometry of Push-Forward Generative Models [24.025975236316846]
  Many deep generative models are defined as a push-forward of a Gaussian measure by a continuous generator, such as Generative Adversarial Networks (GANs) or Variational Auto-Encoders (VAEs) This work explores the latent space of such deep generative models. A key issue with these models is their tendency to output samples outside of the support of the target distribution when learning disconnected distributions.
  arXiv  Detail & Related papers  (2022-07-21T15:29:35Z)
• Adversarial Manifold Matching via Deep Metric Learning for Generative Modeling [5.5840609705075055]
  We propose a manifold matching approach to generative models which includes a distribution generator and a metric generator. The distribution generator aims at generating samples that follow some distribution condensed around the real data manifold. The metric generator utilizes both real data and generated samples to learn a distance metric.
  arXiv  Detail & Related papers  (2021-06-20T23:25:01Z)
• Flow-based Generative Models for Learning Manifold to Manifold Mappings [39.60406116984869]
  We introduce three kinds of invertible layers for manifold-valued data, which are analogous to their functionality in flow-based generative models. We show promising results where we can reliably and accurately reconstruct brain images of a field of orientation distribution functions.
  arXiv  Detail & Related papers  (2020-12-18T02:19:18Z)
• Variational Mixture of Normalizing Flows [0.0]
  Deep generative models, such as generative adversarial networks autociteGAN, variational autoencoders autocitevaepaper, and their variants, have seen wide adoption for the task of modelling complex data distributions. Normalizing flows have overcome this limitation by leveraging the change-of-suchs formula for probability density functions. The present work overcomes this by using normalizing flows as components in a mixture model and devising an end-to-end training procedure for such a model.
  arXiv  Detail & Related papers  (2020-09-01T17:20:08Z)
• Evaluating the Disentanglement of Deep Generative Models through Manifold Topology [66.06153115971732]
  We present a method for quantifying disentanglement that only uses the generative model. We empirically evaluate several state-of-the-art models across multiple datasets.
  arXiv  Detail & Related papers  (2020-06-05T20:54:11Z)

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.