Efficiently Generating Multidimensional Calorimeter Data with Tensor Decomposition Parameterization

Paimon Goulart, Shaan Pakala, Evangelos E. Papalexakis

Keywords: Tensor Decomposition, Generative Models, Efficient Machine Learning, GAN, Diffusion Models

TL;DR: In this work we utilize tensor decomposition to significantly reduce the number of required parameters for generating large multidimensional calorimeter data to increase efficiency.

Abstract: Producing large complex simulation datasets can often be a time and resource consuming task. Especially when these experiments are very expensive, it is becoming more reasonable to generate synthetic data for downstream tasks. Recently, these methods may include using generative machine learning models such as Generative Adversarial Networks or diffusion models. As these generative models improve efficiency in producing useful data, we introduce an internal tensor decomposition to these generative models to even further reduce costs. More specifically, for multidimensional data, or tensors, we generate the smaller tensor factors instead of the full tensor, in order to significantly reduce the model's output parameter, and overall parameters. In doing this, we can significantly reduce the costs of generating complicated simulation data, and we show in our experiments that the generated data can be just as useful. As a result, tensor decomposition has the potential to improve efficiency in generative models, especially when generating multidimensional data, or tensors.

Submission Number: 37