Data for: Foundations of a fast, data-driven, machine-learned simulator
Authors/Creators
- 1. University of California, Irvine
Description
We introduce Optimal-Transport-based Unfolding and Simulation (OTUS), a novel, fast simulator based on unsupervised machine-learning that is capable of predicting experimental data from theoretical models. Simulations are crucial in science because they map from theoretical models to experimental data, allowing scientists to test predictions of theoretical models against the reality of experiments. Experimental data is often reconstructed from indirect measurements causing the aggregate transformation from theoretical models to experimental data to be poorly described by analytical methods. Scientists instead rely on ad-hoc, numerical simulations at great computational cost. Capable of learning directly from data, OTUS trains a probabilistic autoencoder to transform directly between theoretical models and experimental data. This is achieved by identifying the probabilistic autoencoder's latent space with the space of theoretical models, causing the decoder network to become a fast, predictive simulator with the potential to replace current, computationally costly simulators. Using particle physics as an illustrative example, we provide proof-of-principle results for Z-boson and top-quark decays, but stress that OTUS can be widely applied to other fields.
Notes
Further detail can be found in attached readme file (readme_data.txt), the publication (preprint: https://arxiv.org/abs/2101.08944), and the code repository linked with this dataset.
Funding provided by: National Science Foundation
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100000001
Award Number: DGE-1633631
Funding provided by: National Science Foundation
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100000001
Award Number: DGE-1839285
Funding provided by: Office of Science
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100006132
Award Number: DE-SC0009920
Funding provided by: Hasso Plattner Foundation
Crossref Funder Registry ID: http://dx.doi.org/10.13039/501100020408
Award Number:
Funding provided by: National Science Foundation
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100000001
Award Number: 1928718
Funding provided by: National Science Foundation
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100000001
Award Number: 2003237
Funding provided by: National Science Foundation
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100000001
Award Number: 2007719
Funding provided by: Intel Corporation
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100002418
Award Number:
Funding provided by: Qualcomm
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100005144
Award Number:
Funding provided by: Defense Advanced Research Projects Agency
Crossref Funder Registry ID: http://dx.doi.org/10.13039/100000185
Award Number: HR001120C0021
Files
readme_data.txt
Files
(241.0 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:2382fcd52cb52de765126c3db994a921
|
162.3 MB | Download |
|
md5:3f02f5c92adb485597e51d7e4e479e2e
|
78.7 MB | Download |
|
md5:8ca5d8e95c3fe74439f5704be275671c
|
6.3 kB | Preview Download |
Additional details
Related works
- Is cited by
- https://arxiv.org/abs/2101.08944 (URL)
- Is derived from
- 10.5281/zenodo.4706055 (DOI)
- 10.5281/zenodo.4706191 (DOI)