Compositional Sculpting of Iterative Generative Processes

Timur Garipov, Sebastiaan De Peuter, Ge Yang, Vikas Gary, Samuel Kaski, Tommi Jaakkola

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

model reuse and composition. A key challenge in composing iterative generative processes, such as GFlowNets and diffusion models, is that to realize the desired target distribution, all steps of the generative process need to be coordinated, and satisfy delicate balance conditions. In this work, we propose Compositional Sculpting: a general approach for defining compositions of iterative generative processes. We then introduce a method for sampling from these compositions built on classifier guidance. We showcase ways to accomplish compositional sculpting in both GFlowNets and diffusion models. We highlight two binary operations - the harmonic mean and the contast - between pairs, and the generalization of these operations to multiple component distributions. We offer empirical results on image and molecular generation tasks. Project codebase: https://github.com/timgaripov/compositional-sculpting.
Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 36 (NeurIPS 2023)
Place of PublicationNew Orleans, United States
Publication statusPublished - 10 Dec 2023

Keywords

  • generative models
  • generative process

Research Beacons, Institutes and Platforms

  • Christabel Pankhurst Institute
  • Institute for Data Science and AI
  • Digital Futures

Fingerprint

Dive into the research topics of 'Compositional Sculpting of Iterative Generative Processes'. Together they form a unique fingerprint.

Cite this