Fast inference of Boosted Decision Trees in FPGAs for particle physics
We describe the implementation of Boosted Decision Trees in the hls4ml library, which allows the translation of a trained model into FPGA firmware through an automated conversion process. Thanks to its fully on-chip implementation, hls4ml performs inference of Boosted Decision Tree models with extre...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2021
|
| Online Access: | https://hdl.handle.net/1721.1/134045 |
| _version_ | 1826193753580240896 |
|---|---|
| author | Summers, S Guglielmo, G Di Duarte, J Harris, P Hoang, D Jindariani, S Kreinar, E Loncar, V Ngadiuba, J Pierini, M Rankin, D Tran, N Wu, Z |
| author2 | Massachusetts Institute of Technology. Department of Physics |
| author_facet | Massachusetts Institute of Technology. Department of Physics Summers, S Guglielmo, G Di Duarte, J Harris, P Hoang, D Jindariani, S Kreinar, E Loncar, V Ngadiuba, J Pierini, M Rankin, D Tran, N Wu, Z |
| author_sort | Summers, S |
| collection | MIT |
| description | We describe the implementation of Boosted Decision Trees in the hls4ml library, which allows the translation of a trained model into FPGA firmware through an automated conversion process. Thanks to its fully on-chip implementation, hls4ml performs inference of Boosted Decision Tree models with extremely low latency. With a typical latency less than 100 ns, this solution is suitable for FPGA-based real-time processing, such as in the Level-1 Trigger system of a collider experiment. These developments open up prospects for physicists to deploy BDTs in FPGAs for identifying the origin of jets, better reconstructing the energies of muons, and enabling better selection of rare signal processes. |
| first_indexed | 2024-09-23T09:44:22Z |
| format | Article |
| id | mit-1721.1/134045 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T09:44:22Z |
| publishDate | 2021 |
| publisher | IOP Publishing |
| record_format | dspace |
| spelling | mit-1721.1/1340452023-03-01T21:42:09Z Fast inference of Boosted Decision Trees in FPGAs for particle physics Summers, S Guglielmo, G Di Duarte, J Harris, P Hoang, D Jindariani, S Kreinar, E Loncar, V Ngadiuba, J Pierini, M Rankin, D Tran, N Wu, Z Massachusetts Institute of Technology. Department of Physics We describe the implementation of Boosted Decision Trees in the hls4ml library, which allows the translation of a trained model into FPGA firmware through an automated conversion process. Thanks to its fully on-chip implementation, hls4ml performs inference of Boosted Decision Tree models with extremely low latency. With a typical latency less than 100 ns, this solution is suitable for FPGA-based real-time processing, such as in the Level-1 Trigger system of a collider experiment. These developments open up prospects for physicists to deploy BDTs in FPGAs for identifying the origin of jets, better reconstructing the energies of muons, and enabling better selection of rare signal processes. 2021-10-27T19:57:47Z 2021-10-27T19:57:47Z 2020 2020-11-20T14:20:18Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/134045 en 10.1088/1748-0221/15/05/P05026 Journal of Instrumentation Creative Commons Attribution 3.0 unported license https://creativecommons.org/licenses/by/3.0/ application/pdf IOP Publishing IOP Publishing |
| spellingShingle | Summers, S Guglielmo, G Di Duarte, J Harris, P Hoang, D Jindariani, S Kreinar, E Loncar, V Ngadiuba, J Pierini, M Rankin, D Tran, N Wu, Z Fast inference of Boosted Decision Trees in FPGAs for particle physics |
| title | Fast inference of Boosted Decision Trees in FPGAs for particle physics |
| title_full | Fast inference of Boosted Decision Trees in FPGAs for particle physics |
| title_fullStr | Fast inference of Boosted Decision Trees in FPGAs for particle physics |
| title_full_unstemmed | Fast inference of Boosted Decision Trees in FPGAs for particle physics |
| title_short | Fast inference of Boosted Decision Trees in FPGAs for particle physics |
| title_sort | fast inference of boosted decision trees in fpgas for particle physics |
| url | https://hdl.handle.net/1721.1/134045 |
| work_keys_str_mv | AT summerss fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT guglielmogdi fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT duartej fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT harrisp fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT hoangd fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT jindarianis fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT kreinare fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT loncarv fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT ngadiubaj fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT pierinim fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT rankind fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT trann fastinferenceofboosteddecisiontreesinfpgasforparticlephysics AT wuz fastinferenceofboosteddecisiontreesinfpgasforparticlephysics |