SciPost Submission Page
Bayesian Active Search on Parameter Space: a 95 GeV Spin-0 Resonance in the ($B-L$)SSM
by Mauricio A. Diaz, Giorgio Cerro, Srinandan Dasmahapatra, Stefano Moretti
This is not the latest submitted version.
Submission summary
| Authors (as registered SciPost users): | Mauricio A. Diaz |
| Submission information | |
|---|---|
| Preprint Link: | https://arxiv.org/abs/2404.18653v1 (pdf) |
| Code repository: | https://github.com/mjadiaz/blssm-bcastor |
| Date submitted: | May 31, 2025, 3:41 p.m. |
| Submitted by: | Mauricio A. Diaz |
| Submitted to: | SciPost Physics |
| Ontological classification | |
|---|---|
| Academic field: | Physics |
| Specialties: |
|
| Approaches: | Computational, Phenomenological |
Abstract
In the attempt to explain possible data anomalies from collider experiments in terms of New Physics (NP) models, computationally expensive scans over their parameter spaces are typically required in order to match theoretical predictions to experimental observations. Under the assumption that anomalies seen at a mass of about 95 GeV by the Large Electron-Positron (LEP) and Large Hadron Collider (LHC) experiments correspond to a NP signal, which we attempt to interpret as a spin-0 resonance in the $(B-L)$ Supersymmetric Standard Model ($(B-L)$SSM), chosen as an illustrative example, we introduce a novel Machine Learning (ML) approach based on a multi-objective active search method, called b-CASTOR, able to achieve high sample efficiency and diversity, due to the use of probabilistic surrogate models and a volume based search policy, outperforming competing algorithms, such as those based on Markov-Chain Monte Carlo (MCMC) methods.
Author indications on fulfilling journal expectations
- Provide a novel and synergetic link between different research areas.
- Open a new pathway in an existing or a new research direction, with clear potential for multi-pronged follow-up work
- Detail a groundbreaking theoretical/experimental/computational discovery
- Present a breakthrough on a previously-identified and long-standing research stumbling block
Current status:
Reports on this Submission
Report #1 by Anonymous (Referee 1) on 2025-7-9 (Invited Report)
- Cite as: Anonymous, Report on arXiv:2404.18653v1, delivered 2025-07-09, doi: 10.21468/SciPost.Report.11511
Strengths
2 - algorithm significantly outperforms the benchmark
3 - associated code is public and well documented
Weaknesses
2 - public code is not linked in the paper
Report
The paper is well written. The physics application is explained in a self contained manner. The algorithm is introduced with a clear notation. A Python implementation is publicly available in a GitHub repository, which is useful and well documented; unfortunately it is not linked in the paper. The multidimensional results are displayed in well-designed figures which clearly demonstrate the efficiency of the new algorithm. The conclusions explore avenues of improving on b-CASTOR and the benchmarks. This could be extended with an outlook of which BSM models could most benefit from b-CASTOR.
The paper provides a valuable contribution to the task of identifying valid regions in parameter space of BSM models. Its methods can be applied to a range of models and the authors give an outlook for improving the algorithm. I therefore recommend to accept this paper for publication in SciPost after a minor revision (see "Requested Changes").
Requested changes
1 - In Section 2.3, second paragraph, the authors state that they use HiggsBounds and HiggsSignals to test for experimental constraints. Additional constraints could be obtained from recasting tools like MadAnalysis5. The authors should explain why they expect its contributions to be subleading or - if MadAnalysis5 was excluded because of runtime - state this as a limitation.
2 - The title of Section 3.4 "Performance study" seems unfitting since no study is performed in this section. Perhaps "Benchmark algorithm" or something similar might be clearer.
3 - The public code to reproduce the paper's results should be linked in Section 3.5. That is both the repo mjadiaz/blssm-bcastor and the implementation of bcastor in mjadiaz/hep-aid.
4 - It seems that the algorithm presented in the paper is applicable to a vast range of BSM models. This should be stated in the conclusions. It would be particularly useful if the authors could highlight BSM models that can most benefit from a search with b-CASTOR.
Typos:
5 - Section 1, paragraph "This work aims at filling...", first sentence change "being out benchmark" to "being our benchmark"
6 - Section 2.2, first paragraph change "constants are denoted as $y_v$" to "constants are denoted as $y_\nu$"
7 - Section 2.2, second paragraph last sentence change $SU(2)_I$ to $SU(2)_L$
8 - Section 4.2, first paragraph last sentence change "In this work, for we discard" to "In this work, we discard"
Recommendation
Ask for minor revision
We sincerely thank the Editor and both referees for their careful reading and constructive feed-back. We have revised the manuscript accordingly and made relevant changes for a resubmission to a different journal option on SciPost, following the Editor's recommendation.
Author: Mauricio A. Diaz on 2025-09-16 [id 5829]
(in reply to Report 2 on 2025-07-11)We sincerely thank the Editor and both referees for their careful reading and constructive feed-back. We have revised the manuscript accordingly and made relevant changes for a resubmission to a different journal option on SciPost, following the Editor's recommendation.