Real-time artificial intelligence for accelerator control: A study at the Fermilab Booster

John, J. St.; Herwig, T. C.; Kafkes, Diana; Mitrevski, J.; Pellico, W.; Perdue, G. N.; Quintero-Parra, Andres; A., Schupbach, B.; Seiya, K.; Tran, N. V.; Schram, M.; Duarte, J.; Huang, Yunzhi; Keller, Rachael T.

doi:10.1103/physrevaccelbeams.24.104601

Cited by 27 publications

(10 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reinforcement learning [16,17] recently achieved impressive results in many domains of applied research, such as robotics [18], self-driving cars [19], gaming [20]. Coming to high energy physics, it has been mainly suggested for jets reconstruction [21,22] and on-line control system for accelerator machines [23]. In this section we discuss the possibility to train a RL agent to play the role of the devil's advocate (RL Advocate).…”

Section: Agentmentioning

confidence: 99%

“…In recent years, different experiments [28,29,30,31] measured a deviation in one of the angular observables of B 0 → K * 0 µ + µ − decays, namely P 5 , in a region of q 2 between 4 and 8 GeV 2 , where q 2 is defined as the di-muon invariant mass squared. In the following, we will use the example of the LHCb measurement of P 5 in the [4.0, 6.0] GeV 2 q 2 bin [28] P 5 = −0.439 ± 0.117 (23) which is found to be higher than the Standard Model value by a factor of 30-40%, depending on the considered theoretical predictions [32,33,34,35].…”

Section: The P 5 Measurementmentioning

confidence: 99%

See 1 more Smart Citation

The DL Advocate: Playing the devil's advocate with hidden systematic uncertainties

Golutvin¹,

Aleksandr²,

Mauri³

et al. 2023

Preprint

View full text Add to dashboard Cite

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, one relies on a combination of gradient descent and optimisation techniques, the other employs reinforcement learning. We illustrate the potentiality of the presented method by considering two examples, firstly the case of a branching fraction measurement of a particle decay and secondly the determination of the P 5 angular observable in 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.

show abstract

Section: Agentmentioning

confidence: 99%

Section: The P 5 Measurementmentioning

confidence: 99%

The DL Advocate: Playing the devil's advocate with hidden systematic uncertainties

Golutvin¹,

Aleksandr²,

Mauri³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…There are many other examples for the use of RL at accelerator facilities, for example, Refs. [240][241][242][243].…”

Section: Model State Environmentmentioning

confidence: 99%

Data-driven science and machine learning methods in laser–plasma physics

Döpp

Eberle

Howard

et al. 2023

High Pow Laser Sci Eng

View full text Add to dashboard Cite

Laser-plasma physics has developed rapidly over the past few decades as lasers have become both more powerful and more widely available. Early experimental and numerical research in this field was dominated by single-shot experiments with limited parameter exploration. However, recent technological improvements make it possible to gather data for hundreds or thousands of different settings in both experiments and simulations. This has sparked interest in using advanced techniques from mathematics, statistics and computer science to deal with, and benefit from, big data. At the same time, sophisticated modeling techniques also provide new ways for researchers to deal effectively with situation where still only sparse data are available. This paper aims to present an overview of relevant machine learning methods with focus on applicability to laser-plasma physics and its important sub-fields of laser-plasma acceleration and inertial confinement fusion.

show abstract

“…Bayesian optimization (and methods based on it), unlike ES, constructs a probabilistic estimate of the unknown functions, in the form of a Gaussian Process (GP), and determines a new point to sample based the fitted function. Neural networks (NN) have been used as surrogate models for magnet control [30] and for simulation-based optimization studies [31]. Neural networks are also being used for uncertainty aware anomaly detection to predict errant beam pulses [32], as virtual diagnostics for 4D tomographic phase space reconstructions [33], for predicting the transverse emittance of space charge dominated beams In Sections IV-B and IV-C we tune several components in this section of the accelerator.…”

Section: B Accelerator Tuning and Optimizationmentioning

confidence: 99%

Practically Safe Extremum Seeking

Williams

Krstić

Scheinker

2022

2022 IEEE 61st Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

We demonstrate the recent designs of Safe Extremum Seeking (Safe ES) on the 1 kilometer-long charged particle accelerator at the Los Alamos Neutron Science Center (LANSCE). Safe ES is a modification of ES which, in addition to minimizing an analytically unknown cost, also employs a safety filter based on an analytically unknown control barrier function (CBF) safety metric.Accelerator tuning is necessitated by the accelerators being large, with many drifting parameters due to thermal effects and degradation. At the same time, safe operation (the maintenance of state constraints) is crucial, as damage brings astronomical costs, both financially and in operation downtime.Our measured (but analytically unknown) safety metric is the beam current. We perform multivariable Safe ES on three accelerator applications, in which we adapt 4, 6, and 3 magnet strength parameters, respectively. Two of the three applications are for validated simulation models of beamlines at LANSCE: the first for the Proton Radiography (pRad) beamline of 800 MeV protons for spot size tuning; the second on a high performance code, HPSim, for tuning the low energy beam transport (LEBT) region of of 750 keV protons. The third is an experimental tuning of the steering magnets in the LEBT at LANSCE.

show abstract

Real-time artificial intelligence for accelerator control: A study at the Fermilab Booster

Cited by 27 publications

References 39 publications

The DL Advocate: Playing the devil's advocate with hidden systematic uncertainties

The DL Advocate: Playing the devil's advocate with hidden systematic uncertainties

Data-driven science and machine learning methods in laser–plasma physics

Practically Safe Extremum Seeking

Contact Info

Product

Resources

About