The DL Advocate: playing the devil’s advocate with hidden systematic uncertainties
Abstract We propose a new method based on machine learning to play the devil’s advocate and investigate the impact of unknown systematic effects in a quantitative way. This method proceeds by reversing the measurement process and using the physics results to interpret systematic effects under the St...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-09-01
|
| Series: | European Physical Journal C: Particles and Fields |
| Online Access: | https://doi.org/10.1140/epjc/s10052-023-11925-w |
| _version_ | 1827781070305624064 |
|---|---|
| author | Andrei Golutvin Aleksandr Iniukhin Andrea Mauri Patrick Owen Nicola Serra Andrey Ustyuzhanin |
| author_facet | Andrei Golutvin Aleksandr Iniukhin Andrea Mauri Patrick Owen Nicola Serra Andrey Ustyuzhanin |
| author_sort | Andrei Golutvin |
| collection | DOAJ |
| description | Abstract We propose a new method based on machine learning to play the devil’s advocate and investigate the impact of unknown systematic effects in a quantitative way. This method proceeds by reversing the measurement process and using the physics results to interpret systematic effects under the Standard Model hypothesis. We explore this idea with two alternative approaches: the first one relies on a combination of gradient descent and optimisation techniques, its application and potentiality is illustrated with an example that studies the branching fraction measurement of a heavy-flavour decay. The second method employs reinforcement learning and it is applied to the determination of the $$P_{5}^{'}$$ P 5 ′ angular observable in $$B^0 \rightarrow K^{*0} {\mu ^+\mu ^-}$$ B 0 → K ∗ 0 μ + μ - decays. We find that for the former, the size of a hypothetical hidden systematic uncertainty strongly depends on the kinematic overlap between the signal and normalisation channel, while the latter is very robust against possible mismodellings of the efficiency. |
| first_indexed | 2024-03-11T15:13:34Z |
| format | Article |
| id | doaj.art-4ba260b0064f480ca12235f558061c55 |
| institution | Directory Open Access Journal |
| issn | 1434-6052 |
| language | English |
| last_indexed | 2024-03-11T15:13:34Z |
| publishDate | 2023-09-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | European Physical Journal C: Particles and Fields |
| spelling | doaj.art-4ba260b0064f480ca12235f558061c552023-10-29T12:33:29ZengSpringerOpenEuropean Physical Journal C: Particles and Fields1434-60522023-09-0183911610.1140/epjc/s10052-023-11925-wThe DL Advocate: playing the devil’s advocate with hidden systematic uncertaintiesAndrei Golutvin0Aleksandr Iniukhin1Andrea Mauri2Patrick Owen3Nicola Serra4Andrey Ustyuzhanin5Imperial College of LondonYandex School of Data AnalysisImperial College of LondonPhysik-Institut, Universität ZürichPhysik-Institut, Universität ZürichConstructor University, Campus Ring 1Abstract We propose a new method based on machine learning to play the devil’s advocate and investigate the impact of unknown systematic effects in a quantitative way. This method proceeds by reversing the measurement process and using the physics results to interpret systematic effects under the Standard Model hypothesis. We explore this idea with two alternative approaches: the first one relies on a combination of gradient descent and optimisation techniques, its application and potentiality is illustrated with an example that studies the branching fraction measurement of a heavy-flavour decay. The second method employs reinforcement learning and it is applied to the determination of the $$P_{5}^{'}$$ P 5 ′ angular observable in $$B^0 \rightarrow K^{*0} {\mu ^+\mu ^-}$$ B 0 → K ∗ 0 μ + μ - decays. We find that for the former, the size of a hypothetical hidden systematic uncertainty strongly depends on the kinematic overlap between the signal and normalisation channel, while the latter is very robust against possible mismodellings of the efficiency.https://doi.org/10.1140/epjc/s10052-023-11925-w |
| spellingShingle | Andrei Golutvin Aleksandr Iniukhin Andrea Mauri Patrick Owen Nicola Serra Andrey Ustyuzhanin The DL Advocate: playing the devil’s advocate with hidden systematic uncertainties European Physical Journal C: Particles and Fields |
| title | The DL Advocate: playing the devil’s advocate with hidden systematic uncertainties |
| title_full | The DL Advocate: playing the devil’s advocate with hidden systematic uncertainties |
| title_fullStr | The DL Advocate: playing the devil’s advocate with hidden systematic uncertainties |
| title_full_unstemmed | The DL Advocate: playing the devil’s advocate with hidden systematic uncertainties |
| title_short | The DL Advocate: playing the devil’s advocate with hidden systematic uncertainties |
| title_sort | dl advocate playing the devil s advocate with hidden systematic uncertainties |
| url | https://doi.org/10.1140/epjc/s10052-023-11925-w |
| work_keys_str_mv | AT andreigolutvin thedladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT aleksandriniukhin thedladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT andreamauri thedladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT patrickowen thedladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT nicolaserra thedladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT andreyustyuzhanin thedladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT andreigolutvin dladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT aleksandriniukhin dladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT andreamauri dladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT patrickowen dladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT nicolaserra dladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties AT andreyustyuzhanin dladvocateplayingthedevilsadvocatewithhiddensystematicuncertainties |