{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/recurrent-inference-machines-for-solving","title":"Recurrent Inference Machines for Solving Inverse Problems","arxiv_id":"1706.04008","date":"2017-06-13","proceeding":null,"authors":["Patrick Putzky","Max Welling"],"abstract":"Much of the recent research on solving iterative inference problems focuses\non moving away from hand-chosen inference algorithms and towards learned\ninference. In the latter, the inference process is unrolled in time and\ninterpreted as a recurrent neural network (RNN) which allows for joint learning\nof model and inference parameters with back-propagation through time. In this\nframework, the RNN architecture is directly derived from a hand-chosen\ninference algorithm, effectively limiting its capabilities. We propose a\nlearning framework, called Recurrent Inference Machines (RIM), in which we turn\nalgorithm construction the other way round: Given data and a task, train an RNN\nto learn an inference algorithm. Because RNNs are Turing complete [1, 2] they\nare capable to implement any inference algorithm. The framework allows for an\nabstraction which removes the need for domain knowledge. We demonstrate in\nseveral image restoration experiments that this abstraction is effective,\nallowing us to achieve state-of-the-art performance on image denoising and\nsuper-resolution tasks and superior across-task generalization.","url_abs":"http://arxiv.org/abs/1706.04008v1","url_pdf":"http://arxiv.org/pdf/1706.04008v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"recurrent-inference-machines-for-solving","repo_url":"https://github.com/MLI-lab/Robustness-CS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"recurrent-inference-machines-for-solving","repo_url":"https://github.com/georgeyiasemis/RIM-PyTorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"recurrent-inference-machines-for-solving","repo_url":"https://github.com/pputzky/invertible_rim","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"recurrent-inference-machines-for-solving","repo_url":"https://github.com/ys-koshelev/lgd","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"image-denoising","task_name":"Image Denoising"},{"task_slug":"image-restoration","task_name":"Image Restoration"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.04008","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}