EUVIMEDAtrasti EUVIMED
Open access · CC BY 4.0

New machine learning techniques for simulation-based inference: InferoStatic nets, kernel score estimation, and kernel likelihood ratio estimation

Kyoungchul Kong, Konstantin T. Matchev, Stephen Mrenna, Prasanth Shyamsundar

SciPost Physics Codebases · 2023-07-07

DOI: 10.21468/scipostphyscodeb.14 ↗

At a glance

Journal article in SciPost Physics Codebases, published online on 2023-07-07. Written by Kyoungchul Kong and 3 co-authors. Cited 5 times so far (Crossref); 19 references. The full text is freely available (open access, CC BY 4.0). Funded by: United States Department of Energy.

SantraukaBibliografiniai duomenysCiteRelated studies

Abstract (original language)

We propose an intuitive, machine-learning approach to multiparameter inference, dubbed the InferoStatic Networks (ISN) method, to model the score and likelihood ratio estimators in cases when the probability density can be sampled but not computed directly. The ISN uses a backend neural network that models a scalar function called the inferostatic potential \varphi φ . In addition, we introduce new strategies, respectively called Kernel Score Estimation (KSE) and Kernel Likelihood Ratio Estimation (KLRE), to learn the score and the likelihood ratio functions from simulated data. We illustrate the new techniques with some toy examples and compare to existing approaches in the literature. We mention en passant some new loss functions that optimally incorporate latent information from simulations into the training procedure.

Funding

  • United States Department of Energy

Cite

APA

Kong, K., Matchev, K. T., Mrenna, S., & Shyamsundar, P. (2023). New machine learning techniques for simulation-based inference: InferoStatic nets, kernel score estimation, and kernel likelihood ratio estimation. SciPost Physics Codebases. https://doi.org/10.21468/scipostphyscodeb.14

Vancouver

Kong K, Matchev KT, Mrenna S, Shyamsundar P. New machine learning techniques for simulation-based inference: InferoStatic nets, kernel score estimation, and kernel likelihood ratio estimation. SciPost Physics Codebases. 2023. doi:10.21468/scipostphyscodeb.14

BibTeX
@article{Kong2023,
  author = {Kong, Kyoungchul and Matchev, Konstantin T. and Mrenna, Stephen and Shyamsundar, Prasanth},
  title = {New machine learning techniques for simulation-based inference: InferoStatic nets, kernel score estimation, and kernel likelihood ratio estimation},
  journal = {SciPost Physics Codebases},
  year = {2023},
  doi = {10.21468/scipostphyscodeb.14}
}
Start for free →