Sparse-to-Sparse Training of Diffusion Models

• URL: http://arxiv.org/abs/2504.21380v1
• Date: Wed, 30 Apr 2025 07:28:11 GMT
• Title: Sparse-to-Sparse Training of Diffusion Models
• Authors: Inês Cardoso Oliveira, Decebal Constantin Mocanu, Luis A. Leiva,
• Abstract summary: This paper introduces, for the first time, the paradigm of sparse-to-sparse training to DMs.<n>We focus on unconditional generation and train sparse DMs from scratch on six datasets.<n>Our experiments show that sparse DMs are able to match and often outperform their counterparts, while substantially reducing the number of trainable parameters and FLOPs.
• Score: 13.443846454835867
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Diffusion models (DMs) are a powerful type of generative models that have achieved state-of-the-art results in various image synthesis tasks and have shown potential in other domains, such as natural language processing and temporal data modeling. Despite their stable training dynamics and ability to produce diverse high-quality samples, DMs are notorious for requiring significant computational resources, both in the training and inference stages. Previous work has focused mostly on increasing the efficiency of model inference. This paper introduces, for the first time, the paradigm of sparse-to-sparse training to DMs, with the aim of improving both training and inference efficiency. We focus on unconditional generation and train sparse DMs from scratch (Latent Diffusion and ChiroDiff) on six datasets using three different methods (Static-DM, RigL-DM, and MagRan-DM) to study the effect of sparsity in model performance. Our experiments show that sparse DMs are able to match and often outperform their Dense counterparts, while substantially reducing the number of trainable parameters and FLOPs. We also identify safe and effective values to perform sparse-to-sparse training of DMs.

Related papers

• Pruning then Reweighting: Towards Data-Efficient Training of Diffusion Models [33.09663675904689]
  We investigate efficient diffusion training from the perspective of dataset pruning. Inspired by the principles of data-efficient training for generative models such as generative adversarial networks (GANs), we first extend the data selection scheme used in GANs to DM training. To further improve the generation performance, we employ a class-wise reweighting approach.
  arXiv  Detail & Related papers  (2024-09-27T20:21:19Z)
• DKDM: Data-Free Knowledge Distillation for Diffusion Models with Any Architecture [69.58440626023541]
  Diffusion models (DMs) have demonstrated exceptional generative capabilities across various domains.<n>DMs now consume increasingly large amounts of data.<n>We propose a novel scenario: using existing DMs as data sources to train new DMs with any architecture.
  arXiv  Detail & Related papers  (2024-09-05T14:12:22Z)
• DANCE: Dual-View Distribution Alignment for Dataset Condensation [39.08022095906364]
  We propose a new DM-based method named Dual-view distribution AligNment for dataset CondEnsation (DANCE) Specifically, from the inner-class view, we construct multiple "middle encoders" to perform pseudo long-term distribution alignment. While from the inter-class view, we use the expert models to perform distribution calibration.
  arXiv  Detail & Related papers  (2024-06-03T07:22:17Z)
• Slight Corruption in Pre-training Data Makes Better Diffusion Models [71.90034201302397]
  Diffusion models (DMs) have shown remarkable capabilities in generating high-quality images, audios, and videos. DMs benefit significantly from extensive pre-training on large-scale datasets. However, pre-training datasets often contain corrupted pairs where conditions do not accurately describe the data. This paper presents the first comprehensive study on the impact of such corruption in pre-training data of DMs.
  arXiv  Detail & Related papers  (2024-05-30T21:35:48Z)
• Towards Faster Training of Diffusion Models: An Inspiration of A Consistency Phenomenon [16.416356358224842]
  Diffusion models (DMs) are a powerful generative framework that have attracted significant attention in recent years. We propose two strategies to accelerate the training of DMs.
  arXiv  Detail & Related papers  (2024-03-14T13:27:04Z)
• Diffusion-Based Neural Network Weights Generation [80.89706112736353]
  D2NWG is a diffusion-based neural network weights generation technique that efficiently produces high-performing weights for transfer learning. Our method extends generative hyper-representation learning to recast the latent diffusion paradigm for neural network weights generation. Our approach is scalable to large architectures such as large language models (LLMs), overcoming the limitations of current parameter generation techniques.
  arXiv  Detail & Related papers  (2024-02-28T08:34:23Z)
• Adaptive Training Meets Progressive Scaling: Elevating Efficiency in Diffusion Models [52.1809084559048]
  We propose a novel two-stage divide-and-conquer training strategy termed TDC Training. It groups timesteps based on task similarity and difficulty, assigning highly customized denoising models to each group, thereby enhancing the performance of diffusion models. While two-stage training avoids the need to train each model separately, the total training cost is even lower than training a single unified denoising model.
  arXiv  Detail & Related papers  (2023-12-20T03:32:58Z)
• Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learning [101.66860222415512]
  Multi-Task Diffusion Model (textscMTDiff) is a diffusion-based method that incorporates Transformer backbones and prompt learning for generative planning and data synthesis. For generative planning, we find textscMTDiff outperforms state-of-the-art algorithms across 50 tasks on Meta-World and 8 maps on Maze2D.
  arXiv  Detail & Related papers  (2023-05-29T05:20:38Z)
• 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)

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.