Projects per year
Abstract
High training costs of generative models and the need to fine-tune them for specific tasks have created a strong interest in 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 language | English |
---|---|
Title of host publication | Advances in Neural Information Processing Systems 36 - 37th Conference on Neural Information Processing Systems, NeurIPS 2023 |
Publisher | Curran Associates Inc. |
Pages | 12665-12702 |
ISBN (Electronic) | 978-1-7138-9992-1 |
Publication status | Published - 2024 |
MoE publication type | A4 Conference publication |
Event | Conference on Neural Information Processing Systems - Ernest N. Morial Convention Center, New Orleans, United States Duration: 10 Dec 2023 → 16 Dec 2023 Conference number: 37 https://nips.cc/ |
Publication series
Name | Advances in Neural Information Processing Systems |
---|---|
Publisher | Morgan Kaufmann Publishers |
Volume | 36 |
ISSN (Print) | 1049-5258 |
Conference
Conference | Conference on Neural Information Processing Systems |
---|---|
Abbreviated title | NeurIPS |
Country/Territory | United States |
City | New Orleans |
Period | 10/12/2023 → 16/12/2023 |
Internet address |
Fingerprint
Dive into the research topics of 'Compositional Sculpting of Iterative Generative Processes'. Together they form a unique fingerprint.-
HEALED/Garg: Human-steered next-generation machine learning for reviving drug design
Garg, V. (Principal investigator)
01/09/2021 → 31/08/2025
Project: RCF Academy Project
-
HEALED/Kaski S.: Human-steered next-generation machine learning for reviving drug design (HEALED)
Kaski, S. (Principal investigator)
01/09/2021 → 31/08/2025
Project: RCF Academy Project
-
MAMAA /Kaski S.: Maximally Autonomous AI Assistant/Kaski S.
Kaski, S. (Principal investigator)
01/01/2022 → 31/12/2024
Project: RCF Academy Project targeted call