{"id":"arxiv/2603.23106","url":"https://arxiv.org/abs/2603.23106","tags":["stat.ML","cs.LG","math.NA","quant-ph"],"score":0,"title":"High-Resolution Tensor-Network Fourier Methods for Exponentially Compressed Non-Gaussian Aggregate Distributions","domain":"stat.ML","source":"arxiv","authors":["Juan José Rodríguez-Aldavero","Juan José García-Ripoll"],"reviews":0,"subject":"stat.ML","summary":"Characteristic functions of weighted sums of independent random variables exhibit low-rank structure in the quantized tensor train (QTT) representation, also known as matrix product states (MPS), enabling up to exponential compression of their fully non-Gaussian probability distributions. Under variable independence, the global characteristic function factorizes into local terms. Its low-rank QTT structure arises from intrinsic spectral smoothness in continuous models, or from spectral energy concentration as the number of components $D$ grows in discrete models. We demonstrate this on…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-03-24","reactions":{}}