DisCo-Diff: Enhancing Continuous Diffusion Models with Discrete Latents

• URL: http://arxiv.org/abs/2407.03300v1
• Date: Wed, 3 Jul 2024 17:42:46 GMT
• Title: DisCo-Diff: Enhancing Continuous Diffusion Models with Discrete Latents
• Authors: Yilun Xu, Gabriele Corso, Tommi Jaakkola, Arash Vahdat, Karsten Kreis,
• Abstract summary: We propose DisCo-Diff to simplify encoding a complex data distribution into a single continuous Gaussian distribution. DisCo-Diff does not rely on pre-trained networks, making the framework universally applicable. We validate DisCo-Diff on toy data, several image synthesis tasks as well as molecular docking, and find that introducing discrete latents consistently improves model performance.
• Score: 41.86208391836456
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion models (DMs) have revolutionized generative learning. They utilize a diffusion process to encode data into a simple Gaussian distribution. However, encoding a complex, potentially multimodal data distribution into a single continuous Gaussian distribution arguably represents an unnecessarily challenging learning problem. We propose Discrete-Continuous Latent Variable Diffusion Models (DisCo-Diff) to simplify this task by introducing complementary discrete latent variables. We augment DMs with learnable discrete latents, inferred with an encoder, and train DM and encoder end-to-end. DisCo-Diff does not rely on pre-trained networks, making the framework universally applicable. The discrete latents significantly simplify learning the DM's complex noise-to-data mapping by reducing the curvature of the DM's generative ODE. An additional autoregressive transformer models the distribution of the discrete latents, a simple step because DisCo-Diff requires only few discrete variables with small codebooks. We validate DisCo-Diff on toy data, several image synthesis tasks as well as molecular docking, and find that introducing discrete latents consistently improves model performance. For example, DisCo-Diff achieves state-of-the-art FID scores on class-conditioned ImageNet-64/128 datasets with ODE sampler.

Related papers

• DDIL: Improved Diffusion Distillation With Imitation Learning [57.3467234269487]
  Diffusion models excel at generative modeling (e.g., text-to-image) but sampling requires multiple denoising network passes. Progressive distillation or consistency distillation have shown promise by reducing the number of passes. We show that DDIL consistency improves on baseline algorithms of progressive distillation (PD), Latent consistency models (LCM) and Distribution Matching Distillation (DMD2)
  arXiv  Detail & Related papers  (2024-10-15T18:21:47Z)
• Discrete generative diffusion models without stochastic differential equations: a tensor network approach [1.5839621757142595]
  Diffusion models (DMs) are a class of generative machine learning methods. We show how to use networks (TNs) to efficiently define and sample such discrete models''
  arXiv  Detail & Related papers  (2024-07-15T18:00:11Z)
• Complexity Matters: Rethinking the Latent Space for Generative Modeling [65.64763873078114]
  In generative modeling, numerous successful approaches leverage a low-dimensional latent space, e.g., Stable Diffusion. In this study, we aim to shed light on this under-explored topic by rethinking the latent space from the perspective of model complexity.
  arXiv  Detail & Related papers  (2023-07-17T07:12:29Z)
• Semi-Implicit Denoising Diffusion Models (SIDDMs) [50.30163684539586]
  Existing models such as Denoising Diffusion Probabilistic Models (DDPM) deliver high-quality, diverse samples but are slowed by an inherently high number of iterative steps. We introduce a novel approach that tackles the problem by matching implicit and explicit factors. We demonstrate that our proposed method obtains comparable generative performance to diffusion-based models and vastly superior results to models with a small number of sampling steps.
  arXiv  Detail & Related papers  (2023-06-21T18:49:22Z)
• Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models [77.83923746319498]
  We propose a framework called Diff-Instruct to instruct the training of arbitrary generative models. We show that Diff-Instruct results in state-of-the-art single-step diffusion-based models. Experiments on refining GAN models show that the Diff-Instruct can consistently improve the pre-trained generators of GAN models.
  arXiv  Detail & Related papers  (2023-05-29T04:22:57Z)
• Hierarchical Integration Diffusion Model for Realistic Image Deblurring [71.76410266003917]
  Diffusion models (DMs) have been introduced in image deblurring and exhibited promising performance. We propose the Hierarchical Integration Diffusion Model (HI-Diff), for realistic image deblurring. Experiments on synthetic and real-world blur datasets demonstrate that our HI-Diff outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-05-22T12:18:20Z)
• Variational Diffusion Auto-encoder: Latent Space Extraction from Pre-trained Diffusion Models [0.0]
  Variational Auto-Encoders (VAEs) face challenges with the quality of generated images, often presenting noticeable blurriness. This issue stems from the unrealistic assumption that approximates the conditional data distribution, $p(textbfx | textbfz)$, as an isotropic Gaussian. We illustrate how one can extract a latent space from a pre-existing diffusion model by optimizing an encoder to maximize the marginal data log-likelihood.
  arXiv  Detail & Related papers  (2023-04-24T14:44:47Z)
• From Points to Functions: Infinite-dimensional Representations in Diffusion Models [23.916417852496608]
  Diffusion-based generative models learn to iteratively transfer unstructured noise to a complex target distribution. We show that a combination of information content from different time steps gives a strictly better representation for the downstream task.
  arXiv  Detail & Related papers  (2022-10-25T05:30:53Z)
• Generation of data on discontinuous manifolds via continuous stochastic non-invertible networks [6.201770337181472]
  We show how to generate discontinuous distributions using continuous networks. We derive a link between the cost functions and the information-theoretic formulation. We apply our approach to synthetic 2D distributions to demonstrate both reconstruction and generation of discontinuous distributions.
  arXiv  Detail & Related papers  (2021-12-17T17:39:59Z)

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.