Conference item
A continuous time framework for discrete denoising models
- Abstract:
- We provide the first complete continuous time framework for denoising diffusion models of discrete data. This is achieved by formulating the forward noising process and corresponding reverse time generative process as Continuous Time Markov Chains (CTMCs). The model can be efficiently trained using a continuous time version of the ELBO. We simulate the high dimensional CTMC using techniques developed in chemical physics and exploit our continuous time framework to derive high performance samplers that we show can outperform discrete time methods for discrete data. The continuous time treatment also enables us to derive a novel theoretical result bounding the error between the generated sample distribution and the true data distribution.
- Publication status:
- Published
- Peer review status:
- Peer reviewed
Actions
Authors
+ Engineering and Physical Sciences Research Council
More from this funder
- Grant:
- 56726
- EP/R013616/1
- Publisher:
- Curran Associates
- Host title:
- Advances in Neural Information Processing Systems 35 (NeurIPS 2022)
- Volume:
- 37
- Pages:
- 28266-28279
- Publication date:
- 2023-04-01
- Acceptance date:
- 2022-09-14
- Event title:
- 36th Conference on Neural Information Processing Systems (NeurIPS 2022)
- Event location:
- New Orleans, USA
- Event website:
- https://nips.cc/Conferences/2022
- Event start date:
- 2022-11-28
- Event end date:
- 2022-12-09
- ISSN:
-
1049-5258
- EISBN:
- 9781713873129
- ISBN:
- 9781713871088
- Language:
-
English
- Keywords:
- Pubs id:
-
1312346
- Local pid:
-
pubs:1312346
- Deposit date:
-
2022-12-08
Terms of use
- Copyright holder:
- Campbell et al.
- Copyright date:
- 2022
- Rights statement:
- Copyright © (2022) by individual authors and Neural Information Processing Systems Foundation Inc. All rights reserved.
- Notes:
- This is the accepted manuscript version of the paper. The final version is available from the Neural Information Processing Systems Foundation at: https://proceedings.neurips.cc/paper_files/paper/2022/hash/b5b528767aa35f5b1a60fe0aaeca0563-Abstract-Conference.html
If you are the owner of this record, you can report an update to it here: Report update to this record