{"corpus_id":16498180,"paper_sha":"9f24956af6712484ee74efe8c50af6307633ec1e","doi":"10.1016/B978-0-444-63428-3.50180-6","arxiv_id":"1511.05385","pmid":null,"pmcid":null,"mag_id":2273233231,"dblp_id":"journals/corr/UlmasovBCDM15","acl_id":null,"title":"Bayesian Optimization with Dimension Scheduling: Application to Biological Systems","year":2015,"publication_date":"2015-11-17","venue":"arXiv.org","journal":{"name":"ArXiv","pages":null,"volume":"abs/1511.05385"},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle"],"pubmed_pub_types":null,"s2_fields_of_study":["Biology","Mathematics","Computer Science"],"reference_count":19,"citation_count":43,"influential_citation_count":0,"is_open_access":true,"arxiv_categories":["stat.ML","cs.AI","cs.LG","math.OC"],"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/1511.05385","s2_open_access_landing_url":"https://www.semanticscholar.org/paper/9f24956af6712484ee74efe8c50af6307633ec1e","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":"Bayesian Optimization (BO) is a data-efficient method for global black-box optimization of an expensive-to-evaluate fitness function. BO typically assumes that computation cost of BO is cheap, but experiments are time consuming or costly. In practice, this allows us to optimize ten or fewer critical parameters in up to 1,000 experiments. But experiments may be less expensive than BO methods assume: In some simulation models, we may be able to conduct multiple thousands of experiments in a few hours, and the computational burden of BO is no longer negligible compared to experimentation time. To address this challenge we introduce a new Dimension Scheduling Algorithm (DSA), which reduces the computational burden of BO for many experiments. The key idea is that DSA optimizes the fitness function only along a small set of dimensions at each iteration. This DSA strategy (1) reduces the necessary computation time, (2) finds good solutions faster than the traditional BO method, and (3) can be parallelized straightforwardly. We evaluate the DSA in the context of optimizing parameters of dynamic models of microalgae metabolism and show faster convergence than traditional BO.","claims":[{"public_id":"cl_33c0510cae0b858eed8fb1c0d90c6fa7","status":"active","text":"A Dimension Scheduling Algorithm reduces the computational burden of Bayesian optimization by optimizing the fitness function along only a small set of dimensions at each iteration.","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_33c0510cae0b858eed8fb1c0d90c6fa7"},{"public_id":"cl_7b8e1211c83af2c27c684de810cc22f8","status":"active","text":"Dimension Scheduling Algorithm reduces the necessary computation time, finds good solutions faster than traditional Bayesian optimization, and can be parallelized straightforwardly.","confidence":0.97,"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_7b8e1211c83af2c27c684de810cc22f8"},{"public_id":"cl_67eb01dac09516e289f6eaab38a6d08f","status":"active","text":"Dimension Scheduling Algorithm shows faster convergence than traditional Bayesian optimization when optimizing parameters of dynamic models of microalgae metabolism.","confidence":0.96,"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_67eb01dac09516e289f6eaab38a6d08f"}],"concepts":[{"public_id":"co_14e6ce3c85bae3e0c3e7b93a9ce4e6c5","status":"active","name":"convergence","description":"The rate at which the optimization procedure approaches a good solution.","types":["optimization 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_14e6ce3c85bae3e0c3e7b93a9ce4e6c5"},{"public_id":"co_2706ffb5c6bb9d74d5910b185921ebec","status":"active","name":"dynamic models of microalgae metabolism","description":"Time-dependent models describing microalgae metabolic behavior and used as the optimization target.","types":["model"],"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_2706ffb5c6bb9d74d5910b185921ebec"},{"public_id":"co_5b4aa7bff7817b683762bc6384dae23b","status":"active","name":"traditional BO method","description":"The standard Bayesian optimization approach used as a comparison baseline.","types":["baseline"],"aliases":["traditional BO"],"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_5b4aa7bff7817b683762bc6384dae23b"},{"public_id":"co_73ec5c41b46755da766363a6d73db896","status":"active","name":"Bayesian Optimization","description":"A data-efficient method for global black-box optimization of an expensive-to-evaluate fitness function.","types":["method"],"aliases":["BO"],"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_73ec5c41b46755da766363a6d73db896"},{"public_id":"co_85683aa2c25a56c84f113ef9b81d516b","status":"active","name":"computational burden","description":"The amount of computation required by the optimization procedure relative to experimentation time.","types":["cost"],"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_85683aa2c25a56c84f113ef9b81d516b"},{"public_id":"co_9d2571a004aefaf6c3b1ff13d2ed3a89","status":"active","name":"fitness function","description":"The objective function being optimized in the Bayesian optimization setting.","types":["objective function"],"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_9d2571a004aefaf6c3b1ff13d2ed3a89"},{"public_id":"co_a87b3c4ee4aa3ec85935ec8af3f4f295","status":"active","name":"microalgae metabolism","description":"The metabolic system of microalgae considered in the optimization experiments.","types":["biological system"],"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_a87b3c4ee4aa3ec85935ec8af3f4f295"},{"public_id":"co_f006d1678f18fefae1a9c6c74f744320","status":"active","name":"parallelization","description":"The ability to distribute computation across multiple workers or processors.","types":["computational 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_f006d1678f18fefae1a9c6c74f744320"},{"public_id":"co_f0b23a72311a499a623986efa5a45e04","status":"active","name":"Dimension Scheduling Algorithm","description":"A Bayesian optimization strategy that updates only a small set of dimensions at each iteration to reduce computational burden.","types":["method"],"aliases":["DSA"],"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_f0b23a72311a499a623986efa5a45e04"}],"external_ids":{"DOI":"10.1016/B978-0-444-63428-3.50180-6","ArXiv":"1511.05385","PubMed":null,"PubMedCentral":null,"MAG":2273233231,"DBLP":"journals/corr/UlmasovBCDM15","ACL":null},"open_access":{"is_open_access":true,"pdf_url":"http://arxiv.org/pdf/1511.05385","landing_url":"https://www.semanticscholar.org/paper/9f24956af6712484ee74efe8c50af6307633ec1e","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":636272,"paper_uid":"ef090346-7093-4973-91fc-29aae3472aea","canonical_identity":{"paper_id":636272,"paper_uid":"ef090346-7093-4973-91fc-29aae3472aea","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/16498180"}