FlexGP
We describe FlexGP, the first Genetic Programming system to perform symbolic regression on large-scale datasets on the cloud via massive data-parallel ensemble learning. FlexGP provides a decentralized, fault tolerant parallelization framework that runs many copies of Multiple Regression Genetic Pro...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer Netherlands
2016
|
| Online Access: | http://hdl.handle.net/1721.1/103516 |
| _version_ | 1826212062690279424 |
|---|---|
| author | Veeramachaneni, Kalyan Arnaldo, Ignacio Derby, Owen O’Reilly, Una-May |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Veeramachaneni, Kalyan Arnaldo, Ignacio Derby, Owen O’Reilly, Una-May |
| author_sort | Veeramachaneni, Kalyan |
| collection | MIT |
| description | We describe FlexGP, the first Genetic Programming system to perform symbolic regression on large-scale datasets on the cloud via massive data-parallel ensemble learning. FlexGP provides a decentralized, fault tolerant parallelization framework that runs many copies of Multiple Regression Genetic Programming, a sophisticated symbolic regression algorithm, on the cloud. Each copy executes with a different sample of the data and different parameters. The framework can create a fused model or ensemble on demand as the individual GP learners are evolving. We demonstrate our framework by deploying 100 independent GP instances in a massive data-parallel manner to learn from a dataset composed of 515K exemplars and 90 features, and by generating a competitive fused model in less than 10 minutes. |
| first_indexed | 2024-09-23T15:15:51Z |
| format | Article |
| id | mit-1721.1/103516 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T15:15:51Z |
| publishDate | 2016 |
| publisher | Springer Netherlands |
| record_format | dspace |
| spelling | mit-1721.1/1035162022-10-02T01:46:29Z FlexGP Veeramachaneni, Kalyan Arnaldo, Ignacio Derby, Owen O’Reilly, Una-May Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Laboratory for Information and Decision Systems Veeramachaneni, Kalyan Arnaldo, Ignacio Derby, Owen O’Reilly, Una-May We describe FlexGP, the first Genetic Programming system to perform symbolic regression on large-scale datasets on the cloud via massive data-parallel ensemble learning. FlexGP provides a decentralized, fault tolerant parallelization framework that runs many copies of Multiple Regression Genetic Programming, a sophisticated symbolic regression algorithm, on the cloud. Each copy executes with a different sample of the data and different parameters. The framework can create a fused model or ensemble on demand as the individual GP learners are evolving. We demonstrate our framework by deploying 100 independent GP instances in a massive data-parallel manner to learn from a dataset composed of 515K exemplars and 90 features, and by generating a competitive fused model in less than 10 minutes. Li Ka Shing Foundation GE Global Research Center 2016-07-01T20:33:34Z 2016-07-01T20:33:34Z 2014-11 2014-06 2016-05-23T12:07:41Z Article http://purl.org/eprint/type/JournalArticle 1570-7873 1572-9184 http://hdl.handle.net/1721.1/103516 Veeramachaneni, Kalyan et al. “FlexGP: Cloud-Based Ensemble Learning with Genetic Programming for Large Regression Problems.” Journal of Grid Computing 13.3 (2015): 391–407. en http://dx.doi.org/10.1007/s10723-014-9320-9 Journal of Grid Computing Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. Springer Science+Business Media Dordrecht application/pdf Springer Netherlands Springer Netherlands |
| spellingShingle | Veeramachaneni, Kalyan Arnaldo, Ignacio Derby, Owen O’Reilly, Una-May FlexGP |
| title | FlexGP |
| title_full | FlexGP |
| title_fullStr | FlexGP |
| title_full_unstemmed | FlexGP |
| title_short | FlexGP |
| title_sort | flexgp |
| url | http://hdl.handle.net/1721.1/103516 |
| work_keys_str_mv | AT veeramachanenikalyan flexgp AT arnaldoignacio flexgp AT derbyowen flexgp AT oreillyunamay flexgp |