Abstract | ||
---|---|---|
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-autoregressiveapproach 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. In our experiments, NAOMI demonstrates significant improvement in imputation accuracy (reducing average error by 60% compared to autoregressive counterparts) and generalization for long-range sequences. |
Year | Venue | Keywords |
---|---|---|
2019 | ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019) | physical systems,stochastic dynamics,error propagation,motion tracking |
Field | DocType | Volume |
Autoregressive model,Mean squared prediction error,Propagation of uncertainty,Physical system,Algorithm,Artificial intelligence,Imputation (statistics),Match moving,Mathematics,Recursion,Machine learning,Generative model | Journal | 32 |
ISSN | Citations | PageRank |
1049-5258 | 1 | 0.35 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Liu, Yukai | 1 | 1 | 0.35 |
Qi Yu | 2 | 188 | 12.87 |
Zheng, Stephan | 3 | 1 | 1.70 |
Zhan, Eric | 4 | 1 | 0.35 |
Yue, Yisong | 5 | 1 | 0.35 |