Enhancing photon identification with neural network methods

Abstract We investigate photon–pion discrimination in regimes where electromagnetic showers overlap at the scale of calorimeter granularity. Using full detector simulations with fine-grained calorimeter segmentation of approximately <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mn>0.025</mml:mn> <mml:mo>×</mml:mo> <mml:mn>0.025</mml:mn> </mml:mrow> </mml:math> in <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mo stretchy="false">(</mml:mo> <mml:mi>η</mml:mi> <mml:mo>,</mml:mo> <mml:mi>ϕ</mml:mi> <mml:mo stretchy="false">)</mml:mo> </mml:mrow> </mml:math> , we benchmark three approaches: boosted decision trees on shower-shape variables, dense neural networks on the same features, and a ResNet-based convolutional neural network operating directly on calorimeter cell energies. The ResNet significantly outperformed both baseline methods, achieving further gains when augmented with soft scoring and an auxiliary Δ R regression head. Our results demonstrate that residual convolutional architectures, combined with physics-motivated training strategies, can substantially improve photon identification in high-luminosity collider environments in which overlapping electromagnetic showers challenge traditional methods.