Ramachandran, SN and Mukhopadhyay, R and Agarwal, M and Jawahar, CV and Namboodiri, V (2024) Understanding the Generalization of Pretrained Diffusion Models on Out-of-Distribution Data. In: UNSPECIFIED, pp. 14767-14775.
|
PDF
Pro_aaa_con_art_int_38_13_2024 .pdf - Published Version Download (1MB) | Preview |
Abstract
This work tackles the important task of understanding out-of-distribution behavior in two prominent types of generative models, i.e., GANs and Diffusion models. Understanding this behavior is crucial in understanding their broader utility and risks as these systems are increasingly deployed in our daily lives. Our first contribution is demonstrating that diffusion spaces outperform GANs� latent spaces in inverting high-quality OOD images. We also provide a theoretical analysis attributing this to the lack of prior holes in diffusion spaces. Our second significant contribution is to provide a theoretical hypothesis that diffusion spaces can be projected onto a bounded hypersphere, enabling image manipulation through geodesic traversal between inverted images. Our analysis shows that different geodesics share common attributes for the same manipulation, which we leverage to perform various image manipulations. We conduct thorough empirical evaluations to support and validate our claims. Finally, our third and final contribution introduces a novel approach to the few-shot sampling for out-of-distribution data by inverting a few images to sample from the cluster formed by the inverted latents. The proposed technique achieves state-of-the-art results for the few-shot generation task in terms of image quality. Our research underscores the promise of diffusion spaces in out-of-distribution imaging and offers avenues for further exploration. Copyright © 2024, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
Item Type: | Conference Paper |
---|---|
Publication: | Proceedings of the AAAI Conference on Artificial Intelligence |
Publisher: | Association for the Advancement of Artificial Intelligence |
Additional Information: | The copyright for this article belongs to author. |
Keywords: | Artificial intelligence, Daily lives; Diffusion model; Empirical evaluations; Generalisation; Generative model; High quality; Hyper-spheres; Image manipulation; State of the art, Diffusion |
Department/Centre: | Division of Interdisciplinary Sciences > Centre for Nano Science and Engineering |
Date Deposited: | 22 May 2024 04:20 |
Last Modified: | 22 May 2024 04:20 |
URI: | https://eprints.iisc.ac.in/id/eprint/84814 |
Actions (login required)
![]() |
View Item |