Improved off-policy training of diffusion samplers

• URL: http://arxiv.org/abs/2402.05098v3
• Date: Sun, 26 May 2024 18:06:40 GMT
• Title: Improved off-policy training of diffusion samplers
• Authors: Marcin Sendera, Minsu Kim, Sarthak Mittal, Pablo Lemos, Luca Scimeca, Jarrid Rector-Brooks, Alexandre Adam, Yoshua Bengio, Nikolay Malkin,
• Abstract summary: We study the problem of training diffusion models to sample from a distribution with an unnormalized density or energy function. We benchmark several diffusion-structured inference methods, including simulation-based variational approaches and off-policy methods. Our results shed light on the relative advantages of existing algorithms while bringing into question some claims from past work.
• Score: 93.66433483772055
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We study the problem of training diffusion models to sample from a distribution with a given unnormalized density or energy function. We benchmark several diffusion-structured inference methods, including simulation-based variational approaches and off-policy methods (continuous generative flow networks). Our results shed light on the relative advantages of existing algorithms while bringing into question some claims from past work. We also propose a novel exploration strategy for off-policy methods, based on local search in the target space with the use of a replay buffer, and show that it improves the quality of samples on a variety of target distributions. Our code for the sampling methods and benchmarks studied is made public at https://github.com/GFNOrg/gfn-diffusion as a base for future work on diffusion models for amortized inference.

Related papers

• Learned Reference-based Diffusion Sampling for multi-modal distributions [2.1383136715042417]
  We introduce Learned Reference-based Diffusion Sampler (LRDS), a methodology specifically designed to leverage prior knowledge on the location of the target modes. LRDS proceeds in two steps by learning a reference diffusion model on samples located in high-density space regions. We experimentally demonstrate that LRDS best exploits prior knowledge on the target distribution compared to competing algorithms on a variety of challenging distributions.
  arXiv  Detail & Related papers  (2024-10-25T10:23:34Z)
• Amortized Posterior Sampling with Diffusion Prior Distillation [55.03585818289934]
  We propose a variational inference approach to sample from the posterior distribution for solving inverse problems. We show that our method is applicable to standard signals in Euclidean space, as well as signals on manifold.
  arXiv  Detail & Related papers  (2024-07-25T09:53:12Z)
• Understanding Reinforcement Learning-Based Fine-Tuning of Diffusion Models: A Tutorial and Review [63.31328039424469]
  This tutorial provides a comprehensive survey of methods for fine-tuning diffusion models to optimize downstream reward functions. We explain the application of various RL algorithms, including PPO, differentiable optimization, reward-weighted MLE, value-weighted sampling, and path consistency learning.
  arXiv  Detail & Related papers  (2024-07-18T17:35:32Z)
• New algorithms for sampling and diffusion models [0.0]
  We introduce a novel sampling method for known distributions and a new algorithm for diffusion generative models with unknown distributions. Our approach is inspired by the concept of the reverse diffusion process, widely adopted in diffusion generative models.
  arXiv  Detail & Related papers  (2024-06-14T02:30:04Z)
• A Diffusion Model Framework for Unsupervised Neural Combinatorial Optimization [7.378582040635655]
  Current deep learning approaches rely on generative models that yield exact sample likelihoods. This work introduces a method that lifts this restriction and opens the possibility to employ highly expressive latent variable models. We experimentally validate our approach in data-free Combinatorial Optimization and demonstrate that our method achieves a new state-of-the-art on a wide range of benchmark problems.
  arXiv  Detail & Related papers  (2024-06-03T17:55:02Z)
• Consistent Diffusion Meets Tweedie: Training Exact Ambient Diffusion Models with Noisy Data [74.2507346810066]
  Ambient diffusion is a recently proposed framework for training diffusion models using corrupted data. We present the first framework for training diffusion models that provably sample from the uncorrupted distribution given only noisy training data.
  arXiv  Detail & Related papers  (2024-03-20T14:22:12Z)
• Enhancing Score-Based Sampling Methods with Ensembles [0.0]
  We introduce the underlying methodology, emphasizing its relationship with generative diffusion models and the previously introduced F"ollmer sampler. We demonstrate the efficacy of ensemble strategies through various examples, including low- to medium-dimensionality sampling problems. Our findings highlight the potential of ensemble strategies for modeling complex probability distributions in situations where gradients are unavailable.
  arXiv  Detail & Related papers  (2024-01-31T01:51:29Z)
• Unsupervised Discovery of Interpretable Directions in h-space of Pre-trained Diffusion Models [63.1637853118899]
  We propose the first unsupervised and learning-based method to identify interpretable directions in h-space of pre-trained diffusion models. We employ a shift control module that works on h-space of pre-trained diffusion models to manipulate a sample into a shifted version of itself. By jointly optimizing them, the model will spontaneously discover disentangled and interpretable directions.
  arXiv  Detail & Related papers  (2023-10-15T18:44:30Z)
• Observation-Guided Diffusion Probabilistic Models [41.749374023639156]
  We propose a novel diffusion-based image generation method called the observation-guided diffusion probabilistic model (OGDM) Our approach reestablishes the training objective by integrating the guidance of the observation process with the Markov chain. We demonstrate the effectiveness of our training algorithm using diverse inference techniques on strong diffusion model baselines.
  arXiv  Detail & Related papers  (2023-10-06T06:29:06Z)
• Score-based Source Separation with Applications to Digital Communication Signals [72.6570125649502]
  We propose a new method for separating superimposed sources using diffusion-based generative models. Motivated by applications in radio-frequency (RF) systems, we are interested in sources with underlying discrete nature. Our method can be viewed as a multi-source extension to the recently proposed score distillation sampling scheme.
  arXiv  Detail & Related papers  (2023-06-26T04:12:40Z)

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.