TY - GEN
T1 - Multiresolution Analysis Using Data-Flow Programming
AU - Chaturvedi, Nilesh
AU - Schuchart, Joseph
AU - Harrison, Robert
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2026
Y1 - 2026
N2 - Multiresolution analysis (MRA) is widely used across scientific disciplines for its ability to represent both local and global data features effectively. Representations in MRA are tree-based with levels corresponding to the scale at which the function is represented. The operations that construct these representations and act on them are irregular and data-dependent and require traversal of the tree. The Template Task Graph (TTG) programming model enables data-dependent task discovery and can represent large-scale task graphs compactly by instantiating only those tasks for which inputs are available. We introduce a framework that implements MRA using TTG. Our work explores the representation of functions, operators, and the application of operators on functions in a multiresolution multiwavelet basis. Our construction tackles a variety of feasibility, efficiency, and numerical challenges in higher dimensions.
AB - Multiresolution analysis (MRA) is widely used across scientific disciplines for its ability to represent both local and global data features effectively. Representations in MRA are tree-based with levels corresponding to the scale at which the function is represented. The operations that construct these representations and act on them are irregular and data-dependent and require traversal of the tree. The Template Task Graph (TTG) programming model enables data-dependent task discovery and can represent large-scale task graphs compactly by instantiating only those tasks for which inputs are available. We introduce a framework that implements MRA using TTG. Our work explores the representation of functions, operators, and the application of operators on functions in a multiresolution multiwavelet basis. Our construction tackles a variety of feasibility, efficiency, and numerical challenges in higher dimensions.
KW - Data-Flow
KW - Multiresolution Analysis
KW - Task Based Programming
KW - Template Task Graphs
UR - https://www.scopus.com/pages/publications/105046899020
U2 - 10.1007/978-3-032-27676-6_13
DO - 10.1007/978-3-032-27676-6_13
M3 - Conference contribution
AN - SCOPUS:105046899020
SN - 9783032276759
T3 - Lecture Notes in Computer Science
SP - 151
EP - 160
BT - Asynchronous Many-Task Systems and Applications - 4th International Workshop, WAMTA 2026, Proceedings
A2 - Diehl, Patrick
A2 - Rampp, Markus
A2 - Laure, Erwin
A2 - Schulz, Martin
PB - Springer Science and Business Media Deutschland GmbH
T2 - 4th International Workshop on Asynchronous Many-Task Systems and Applications, WAMTA 2026
Y2 - 16 February 2026 through 18 February 2026
ER -