{"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/end-to-end-refinement-guided-by-pre-trained","title":"End-to-End Refinement Guided by Pre-trained Prototypical Classifier","arxiv_id":"1805.08698","date":"2018-05-07","proceeding":null,"authors":["Junwen Bai","Zihang Lai","Runzhe Yang","Yexiang Xue","John Gregoire","Carla Gomes"],"abstract":"Many real-world tasks involve identifying patterns from data satisfying\nbackground or prior knowledge. In domains like materials discovery, due to the\nflaws and biases in raw experimental data, the identification of X-ray\ndiffraction patterns (XRD) often requires a huge amount of manual work in\nfinding refined phases that are similar to the ideal theoretical ones.\nAutomatically refining the raw XRDs utilizing the simulated theoretical data is\nthus desirable. We propose imitation refinement, a novel approach to refine\nimperfect input patterns, guided by a pre-trained classifier incorporating\nprior knowledge from simulated theoretical data, such that the refined patterns\nimitate the ideal data. The classifier is trained on the ideal simulated data\nto classify patterns and learns an embedding space where each class is\nrepresented by a prototype. The refiner learns to refine the imperfect patterns\nwith small modifications, such that their embeddings are closer to the\ncorresponding prototypes. We show that the refiner can be trained in both\nsupervised and unsupervised fashions. We further illustrate the effectiveness\nof the proposed approach both qualitatively and quantitatively in a digit\nrefinement task and an X-ray diffraction pattern refinement task in materials\ndiscovery.","url_abs":"http://arxiv.org/abs/1805.08698v2","url_pdf":"http://arxiv.org/pdf/1805.08698v2.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":"end-to-end-refinement-guided-by-pre-trained","repo_url":"https://github.com/JunwenBai/crystal-classification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}