N-BEATS: Neural basis expansion analysis for interpretable time series\n forecasting

We focus on solving the univariate times series point forecasting problem\nusing deep learning. We propose a deep neural architecture based on backward\nand forward residual links and a very deep stack of fully-connected layers. The\narchitecture has a number of desirable properties, being interpretable,\napplicable without modification to a wide array of target domains, and fast to\ntrain. We test the proposed architecture on several well-known datasets,\nincluding M3, M4 and TOURISM competition datasets containing time series from\ndiverse domains. We demonstrate state-of-the-art performance for two\nconfigurations of N-BEATS for all the datasets, improving forecast accuracy by\n11% over a statistical…