eprintid: 10198875 rev_number: 7 eprint_status: archive userid: 699 dir: disk0/10/19/88/75 datestamp: 2024-10-24 15:15:00 lastmod: 2024-10-24 15:15:00 status_changed: 2024-10-24 15:15:00 type: proceedings_section metadata_visibility: show sword_depositor: 699 creators_name: della Maggiora, G creators_name: Croquevielle, LA creators_name: Deshpande, N creators_name: Horsley, H creators_name: Heinis, T creators_name: Yakimovich, A title: Conditional Variational Diffusion Models ispublished: pub divisions: UCL divisions: B02 divisions: C10 divisions: D17 divisions: G93 keywords: Denoising Diffusion Probabilistic Models, Inverse Problems, Generative Models, Super Resolution, Phase Quantification, Variational Methods note: This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions. abstract: Inverse problems aim to determine parameters from observations, a crucial task in engineering and science. Lately, generative models, especially diffusion models, have gained popularity in this area for their ability to produce realistic solutions and their good mathematical properties. Despite their success, an important drawback of diffusion models is their sensitivity to the choice of variance schedule, which controls the dynamics of the diffusion process. Fine-tuning this schedule for specific applications is crucial but time-consuming and does not guarantee an optimal result. We propose a novel approach for learning the schedule as part of the training process. Our method supports probabilistic conditioning on data, provides high-quality solutions, and is flexible, proving able to adapt to different applications with minimum overhead. This approach is tested in two unrelated inverse problems: super-resolution microscopy and quantitative phase imaging, yielding comparable or superior results to previous methods and fine-tuned diffusion models. We conclude that fine-tuning the schedule by experimentation should be avoided because it can be learned during training in a stable way that yields better results. date: 2024-01-16 date_type: published publisher: International Conference on Learning Representations (ICLR) official_url: https://openreview.net/forum?id=YOKnEkIuoi oa_status: green full_text_type: pub language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 2329524 lyricists_name: Horsley, Harry lyricists_id: HHORS59 actors_name: Flynn, Bernadette actors_id: BFFLY94 actors_role: owner full_text_status: public pres_type: paper series: ICLR publication: 12th International Conference on Learning Representations, ICLR 2024 volume: 2024 event_title: 12th International Conference on Learning Representations, ICLR 2024 book_title: 12th International Conference on Learning Representations, ICLR 2024 citation: della Maggiora, G; Croquevielle, LA; Deshpande, N; Horsley, H; Heinis, T; Yakimovich, A; (2024) Conditional Variational Diffusion Models. In: 12th International Conference on Learning Representations, ICLR 2024. International Conference on Learning Representations (ICLR) Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10198875/1/5967_Conditional_Variational_D.pdf