A Multi-Branch Decoder Network Approach to Adaptive Temporal Data Selection and Reconstruction for Big Scientific Simulation Data

Yang Zhang, Hanqi Guo, Lanyu Shang, Dong Wang, Tom Peterka

Research output: Contribution to journalArticlepeer-review


A key challenge in scientific simulation is that the simulation outputs often require intensive I/O and storage space to store the results for effective post hoc analysis. This article focuses on a quality-aware adaptive temporal data selection and reconstruction problem where the goal is to adaptively select simulation data samples at certain key timesteps in situ and reconstruct the discarded samples with quality assurance during post hoc analysis. This problem is motivated by the limitation of current solutions that a significant amount of simulation data samples are either discarded or aggregated during the sampling process, leading to inaccurate modeling of the simulated phenomena. Two unique challenges exist: 1) the sampling decisions have to be made in situ and adapted to the dynamics of the complex scientific simulation data; 2) the reconstruction error must be strictly bounded to meet the application requirement. To address the above challenges, we develop DeepSample, an error-controlled convolutional neural network framework, that jointly integrates a set of coherent multi-branch deep decoders to effectively reconstruct the simulation data with rigorous quality assurance. The results on two real-world scientific simulation applications show that DeepSample significantly outperforms other state-of-the-art methods on both sampling efficiency and reconstructed simulation data quality.

Original languageEnglish (US)
Pages (from-to)1637-1649
Number of pages13
JournalIEEE Transactions on Big Data
Issue number6
StatePublished - Dec 1 2022


  • Adaptation models
  • Adaptive Temporal Data Selection and Reconstruction
  • Adaptive systems
  • Analytical models
  • Big Scientific Simulation Data
  • Computational modeling
  • Data integrity
  • Data models
  • Image reconstruction
  • Multi-branch Decoder Network
  • adaptive temporal data selection and reconstruction
  • Big scientific simulation data
  • multi-branch decoder network

ASJC Scopus subject areas

  • Information Systems and Management
  • Information Systems

Cite this