NAOMI: Non-Autoregressive Multiresolution Sequence Imputation
This addresses error propagation issues in spatiotemporal modeling for applications like motion tracking, though it is an incremental improvement over existing non-autoregressive approaches.
The paper tackled the problem of missing value imputation in long-range spatiotemporal sequences by proposing a non-autoregressive deep generative model, which reduced average prediction error by 60% compared to autoregressive methods.
Missing value imputation is a fundamental problem in spatiotemporal modeling, from motion tracking to the dynamics of physical systems. Deep autoregressive models suffer from error propagation which becomes catastrophic for imputing long-range sequences. In this paper, we take a non-autoregressive approach and propose a novel deep generative model: Non-AutOregressive Multiresolution Imputation (NAOMI) to impute long-range sequences given arbitrary missing patterns. NAOMI exploits the multiresolution structure of spatiotemporal data and decodes recursively from coarse to fine-grained resolutions using a divide-and-conquer strategy. We further enhance our model with adversarial training. When evaluated extensively on benchmark datasets from systems of both deterministic and stochastic dynamics. NAOMI demonstrates significant improvement in imputation accuracy (reducing average prediction error by 60% compared to autoregressive counterparts) and generalization for long range sequences.