Measuring the quality of Synthetic data for use in competitions. Jordon, J., Yoon, J., & van der Schaar, M. 3:pages, June, 2018. _eprint: 1806.11345
Paper abstract bibtex Machine learning has the potential to assist many communities in using the large datasets that are becoming more and more available. Unfortunately, much of that potential is not being realized because it would require sharing data in a way that compromises privacy. In order to overcome this hurdle, several methods have been proposed that generate synthetic data while preserving the privacy of the real data. In this paper we consider a key characteristic that synthetic data should have in order to be useful for machine learning researchers - the relative performance of two algorithms (trained and tested) on the synthetic dataset should be the same as their relative performance (when trained and tested) on the original dataset.
@article{jordon_measuring_2018,
title = {Measuring the quality of {Synthetic} data for use in competitions},
volume = {3},
url = {http://arxiv.org/abs/1806.11345},
abstract = {Machine learning has the potential to assist many communities in using the large datasets that are becoming more and more available. Unfortunately, much of that potential is not being realized because it would require sharing data in a way that compromises privacy. In order to overcome this hurdle, several methods have been proposed that generate synthetic data while preserving the privacy of the real data. In this paper we consider a key characteristic that synthetic data should have in order to be useful for machine learning researchers - the relative performance of two algorithms (trained and tested) on the synthetic dataset should be the same as their relative performance (when trained and tested) on the original dataset.},
author = {Jordon, James and Yoon, Jinsung and van der Schaar, Mihaela},
month = jun,
year = {2018},
note = {\_eprint: 1806.11345},
keywords = {Synthetic data, competitions, metrics, privacy},
pages = {pages},
}
Downloads: 0
{"_id":"XCdG35wRCqoEL5LJE","bibbaseid":"jordon-yoon-vanderschaar-measuringthequalityofsyntheticdataforuseincompetitions-2018","authorIDs":[],"author_short":["Jordon, J.","Yoon, J.","van der Schaar, M."],"bibdata":{"bibtype":"article","type":"article","title":"Measuring the quality of Synthetic data for use in competitions","volume":"3","url":"http://arxiv.org/abs/1806.11345","abstract":"Machine learning has the potential to assist many communities in using the large datasets that are becoming more and more available. Unfortunately, much of that potential is not being realized because it would require sharing data in a way that compromises privacy. In order to overcome this hurdle, several methods have been proposed that generate synthetic data while preserving the privacy of the real data. In this paper we consider a key characteristic that synthetic data should have in order to be useful for machine learning researchers - the relative performance of two algorithms (trained and tested) on the synthetic dataset should be the same as their relative performance (when trained and tested) on the original dataset.","author":[{"propositions":[],"lastnames":["Jordon"],"firstnames":["James"],"suffixes":[]},{"propositions":[],"lastnames":["Yoon"],"firstnames":["Jinsung"],"suffixes":[]},{"propositions":["van","der"],"lastnames":["Schaar"],"firstnames":["Mihaela"],"suffixes":[]}],"month":"June","year":"2018","note":"_eprint: 1806.11345","keywords":"Synthetic data, competitions, metrics, privacy","pages":"pages","bibtex":"@article{jordon_measuring_2018,\n\ttitle = {Measuring the quality of {Synthetic} data for use in competitions},\n\tvolume = {3},\n\turl = {http://arxiv.org/abs/1806.11345},\n\tabstract = {Machine learning has the potential to assist many communities in using the large datasets that are becoming more and more available. Unfortunately, much of that potential is not being realized because it would require sharing data in a way that compromises privacy. In order to overcome this hurdle, several methods have been proposed that generate synthetic data while preserving the privacy of the real data. In this paper we consider a key characteristic that synthetic data should have in order to be useful for machine learning researchers - the relative performance of two algorithms (trained and tested) on the synthetic dataset should be the same as their relative performance (when trained and tested) on the original dataset.},\n\tauthor = {Jordon, James and Yoon, Jinsung and van der Schaar, Mihaela},\n\tmonth = jun,\n\tyear = {2018},\n\tnote = {\\_eprint: 1806.11345},\n\tkeywords = {Synthetic data, competitions, metrics, privacy},\n\tpages = {pages},\n}\n\n","author_short":["Jordon, J.","Yoon, J.","van der Schaar, M."],"key":"jordon_measuring_2018","id":"jordon_measuring_2018","bibbaseid":"jordon-yoon-vanderschaar-measuringthequalityofsyntheticdataforuseincompetitions-2018","role":"author","urls":{"Paper":"http://arxiv.org/abs/1806.11345"},"keyword":["Synthetic data","competitions","metrics","privacy"],"downloads":0},"bibtype":"article","biburl":"https://api.zotero.org/users/3522498/collections/AW3NX4WW/items?key=kdJ5QIjIIc7oy1mYjjz70Rv2&format=bibtex&limit=100","creationDate":"2021-03-14T11:19:06.385Z","downloads":0,"keywords":["synthetic data","competitions","metrics","privacy"],"search_terms":["measuring","quality","synthetic","data","use","competitions","jordon","yoon","van der schaar"],"title":"Measuring the quality of Synthetic data for use in competitions","year":2018,"dataSources":["dwrmKCbrccWf5bf2H"]}