{"corpus_id":2988078,"paper_sha":"21d08abf967d6ce56f72f96d8e87b2288f53a4ed","doi":"10.1007/s40304-018-0127-z","arxiv_id":"1710.00211","pmid":null,"pmcid":null,"mag_id":2952003181,"dblp_id":"journals/corr/abs-1710-00211","acl_id":null,"title":"The Deep Ritz Method: A Deep Learning-Based Numerical Algorithm for Solving Variational Problems","year":2017,"publication_date":"2017-09-30","venue":"Communications in Mathematics and Statistics","journal":{"name":"Communications in Mathematics and Statistics","pages":"1 - 12","volume":"6"},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle"],"pubmed_pub_types":null,"s2_fields_of_study":["Mathematics","Computer Science"],"reference_count":15,"citation_count":1698,"influential_citation_count":115,"is_open_access":true,"arxiv_categories":["cs.LG","stat.ML"],"arxiv_license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":"http://arxiv.org/pdf/1710.00211","s2_open_access_landing_url":"https://www.semanticscholar.org/paper/21d08abf967d6ce56f72f96d8e87b2288f53a4ed","s2_open_access_license":null,"s2_open_access_status":"GREEN","pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"We propose a deep learning-based method, the Deep Ritz Method, for numerically solving variational problems, particularly the ones that arise from partial differential equations. The Deep Ritz Method is naturally nonlinear, naturally adaptive and has the potential to work in rather high dimensions. The framework is quite simple and fits well with the stochastic gradient descent method used in deep learning. We illustrate the method on several problems including some eigenvalue problems.","claims":[{"public_id":"cl_b5889a479eb89af41fd462cf85574e17","status":"active","text":"Deep Ritz Method provides a deep learning-based numerical algorithm for solving variational problems arising from partial differential equations.","confidence":0.98,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_b5889a479eb89af41fd462cf85574e17"},{"public_id":"cl_45d2c7d528cef1523f195cc39c9fabe3","status":"active","text":"The framework is simple and fits well with stochastic gradient descent.","confidence":0.91,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_45d2c7d528cef1523f195cc39c9fabe3"},{"public_id":"cl_b704f7af0dbad0dae3f4d2ad6eab46d5","status":"active","text":"The method is illustrated on several problems, including eigenvalue problems.","confidence":0.89,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_b704f7af0dbad0dae3f4d2ad6eab46d5"},{"public_id":"cl_c54166ff9e280b0c8ee9cf929343198a","status":"active","text":"The method is naturally nonlinear, naturally adaptive, and has potential to work in high dimensions.","confidence":0.93,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_c54166ff9e280b0c8ee9cf929343198a"}],"concepts":[{"public_id":"co_28de5013c98f8329f2e03dafb93b9c1a","status":"active","name":"nonlinear method","description":"A computational approach whose model or solution form is not restricted to linear combinations.","types":["method property"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_28de5013c98f8329f2e03dafb93b9c1a"},{"public_id":"co_35efb6dc8cec0a77698d1f62723cb1c3","status":"active","name":"stochastic gradient descent","description":"An iterative optimization algorithm that updates parameters using noisy gradient estimates from sampled data.","types":["optimization method"],"aliases":["SGD"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_35efb6dc8cec0a77698d1f62723cb1c3"},{"public_id":"co_53d6afdd640536cf529c1a9c0e71fe12","status":"active","name":"energy functional","description":"A scalar functional whose minimization defines the variational formulation used by the method.","types":["mathematical object"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_53d6afdd640536cf529c1a9c0e71fe12"},{"public_id":"co_89ebbd81387cd3e735260acd5e19ab0d","status":"active","name":"adaptive method","description":"A method that can adjust its representation or parameters to the problem at hand.","types":["method property"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_89ebbd81387cd3e735260acd5e19ab0d"},{"public_id":"co_a96ae7ecef2ad7205f36411ce422c850","status":"active","name":"eigenvalue problems","description":"Problems of finding scalar values and corresponding nonzero vectors or functions satisfying an operator equation.","types":["problem class"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_a96ae7ecef2ad7205f36411ce422c850"},{"public_id":"co_c52037a7387cf931b1ed45a72c8c67c2","status":"active","name":"partial differential equations","description":"Equations involving partial derivatives that motivate the variational problems considered here.","types":["equation class"],"aliases":["PDEs"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_c52037a7387cf931b1ed45a72c8c67c2"},{"public_id":"co_d0eace25f93bba7a667747ff54783497","status":"active","name":"high-dimensional problems","description":"Problems defined in spaces with many variables or dimensions.","types":["problem class"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_d0eace25f93bba7a667747ff54783497"},{"public_id":"co_d34a93d3482581f25d2f091e1386bb40","status":"active","name":"Deep Ritz Method","description":"A deep learning-based numerical method for solving variational formulations by minimizing an energy functional.","types":["method"],"aliases":["DRM"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_d34a93d3482581f25d2f091e1386bb40"},{"public_id":"co_e59d5719b6195e96da418b0cbc3238d7","status":"active","name":"variational problems","description":"Optimization problems formulated through the minimization or stationarity of an energy functional.","types":["problem class"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_e59d5719b6195e96da418b0cbc3238d7"},{"public_id":"co_f294cb81a089cbe0f57a13fd77fb0c14","status":"active","name":"deep learning-based numerical algorithm","description":"A numerical procedure that uses deep neural networks as part of the solution strategy.","types":["method"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_f294cb81a089cbe0f57a13fd77fb0c14"}],"external_ids":{"DOI":"10.1007/s40304-018-0127-z","ArXiv":"1710.00211","PubMed":null,"PubMedCentral":null,"MAG":2952003181,"DBLP":"journals/corr/abs-1710-00211","ACL":null},"open_access":{"is_open_access":true,"pdf_url":"http://arxiv.org/pdf/1710.00211","landing_url":"https://www.semanticscholar.org/paper/21d08abf967d6ce56f72f96d8e87b2288f53a4ed","source":"semantic_scholar","pdf_url_source":"semantic_scholar_open_access_pdf","license":null,"status":"GREEN","reason":null},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":true,"full_text_source":"arxiv","count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":637832,"paper_uid":"341d7c09-c921-4231-9b70-6260d8435a68","canonical_identity":{"paper_id":637832,"paper_uid":"341d7c09-c921-4231-9b70-6260d8435a68","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/2988078"}