Orateur
Description
Maximizing the scientific discovery rate in complex modern experiments demands advanced data analysis and real-time control. We present a suite of recently developed Artificial Intelligence (AI) and Machine Learning (ML) applications that can transform the precision and efficiency of experimental work at GANIL.
Our efforts are focused on two critical areas:
- Deep Neural Networks (DNNs) for Enhanced Data Fidelity:
- Precision Spectrometry: DNNs were deployed to solve the highly non-linear, multi-parametric problem of ion trajectory reconstruction in the large-acceptance magnetic spectrometer VAMOS++. By training on theoretically generated ray-tracing data, the network drastically improves the handling of complex fringe-field and acceptance effects, leading to a more accurate determination of relevant quantities.
- Heavy Ion Identification: We implemented DNNs to determine the atomic charge state and atomic number of heavy ions from VAMOS++ focal-plane multi-parametric detector data (e.g., ionization chamber). This approach utilizes semi-supervised learning, leveraging a small fraction of precisely labeled experimental events to effectively classify the majority of unlabeled data, addressing a long-standing challenge in robust event tagging.
-
Light Ion Particle Identification: DNNs were successfully used for robust particle identification in highly-segmented silicon telescopes, significantly improving event clarity in complex reaction channels.
-
Bayesian Optimization (BO) for Real-Time System Control of Multi-Parametric Systems:
- Beam Transport: We demonstrate the effectiveness of the BO technique with an optimization study of the beam optics transport through a quadrupole triplet, confirming its potential for real-time autonomous control of the slow control system like beam transmission, beam extraction efficiency and other.
- Spectrometer control: The BO approach is also envisaged for optimizing ion transport through the High-Resolution Separator (HRS) at the DESIR facility.
These tools collectively deliver unprecedented levels of precision in data analysis and system stability at GANIL, directly enhancing the scientific output of the facility.
