Universal uncertainty estimation for nuclear detector signals with neural networks and ensemble learning

Abstract

Characterizing uncertainty is a common issue in nuclear measurement and has important implications for reliable physical discovery. Traditional methods are either insufficient to cope with the heterogeneous nature of uncertainty or inadequate to perform well with unknown mathematical models. In this paper, we propose using multi-layer convolutional neural networks for empirical uncertainty estimation and feature extraction of nuclear pulse signals. This method is based on deep learning, a recent development of machine learning techniques, which learns the desired mapping function from training data and generalizes to unseen test data. Furthermore, ensemble learning is utilized to estimate the uncertainty originating from trainable parameters of the network and to improve the robustness of the whole model. To evaluate the performance of the proposed method, simulation studies, in comparison with curve fitting, investigate extensive conditions and show its universal applicability. Finally, a case study is made using data from a NICA-MPD electromagnetic calorimeter module exposed to a test beam at DESY, Germany. The uncertainty estimation method successfully detected out-of-distribution samples and also achieved good accuracy in time and energy measurements.

Label-free timing analysis of SiPM-based modularized detectors with physics-constrained deep learning

“…It should be noted that NNs are versatile and able to be integrated with other feature extraction tasks, such as energy estimation [36] and position estimation [12]. Besides, the ability of deep learning to exploit information in noisy conditions indicates possible applications in harsh experimental environment, such as the time-of-flight measurement of laser-accelerated particles with interference of high-level electromagnetic pulses [37].…”

An Ultra-Throughput Boost Method for Gamma-Ray Spectrometers

“…For the energy spectrum, deconvolution methods based on constrained optimization have achieved a high-resolution boost [13,14] and may represent a complementary approach to the above pulse throughput enhancing method to compensate for resolution deterioration. With the rapid development of computing power and complex model equation-solving methods and algorithms, some works have used artificial intelligence methods [15][16][17][18] to restore pileup pulses. However, the pileup effect occurs because a certain number of piled-up pulses are rejected to maintain resolution and cannot be recognized.…”