Data-driven detector signal characterization with constrained bottleneck autoencoders

TL;DR

This paper demonstrates how constrained bottleneck autoencoders can learn detector response models directly from data, enabling accurate parameter estimation, high-fidelity simulation, and denoising even with noisy signals.

Contribution

It introduces a deep learning approach using constrained autoencoders to model unknown detector responses directly from data, bypassing traditional parametric modeling.

Findings

Achieves excellent performance in modeling detector responses from noisy data.

Enables simultaneous parameter estimation, simulation, and denoising.

Effective even with significant signal noise.

Abstract

A common technique in high energy physics is to characterize the response of a detector by means of models tunned to data which build parametric maps from the physical parameters of the system to the expected signal of the detector. When the underlying model is unknown it is difficult to apply this method, and often, simplifying assumptions are made introducing modeling errors. In this article, using a waveform toy model we present how deep learning in the form of constrained bottleneck autoencoders can be used to learn the underlying unknown detector response model directly from data. The results show that excellent performance results can be achieved even when the signals are significantly affected by random noise. The trained algorithm can be used simultaneously to perform estimations on the physical parameters of the model, simulate the detector response with high fidelity and to denoise detector signals.