Ajou University repository

Transition Matrix Representation of Trees with Transposed Convolutionsoa mark
Citations

SCOPUS

0

Citation Export

Publication Year
2022-01-01
Journal
Proceedings of the 2022 SIAM International Conference on Data Mining, SDM 2022
Publisher
Society for Industrial and Applied Mathematics Publications
Citation
Proceedings of the 2022 SIAM International Conference on Data Mining, SDM 2022, pp.154-162
Mesh Keyword
Black box modellingClassification accuracyDesign parametersInterpretabilityMatrix representationPerformanceTransition matricesTree modelingTree representationTree structures
All Science Classification Codes (ASJC)
Computer Science ApplicationsSoftware
Abstract
How can we effectively find the best structures in tree models? Tree models have been favored over complex black box models in domains where interpretability is crucial for making irreversible decisions. However, searching for a tree structure that gives the best balance between the performance and the interpretability remains a challenging task. In this paper, we propose Tart (Transition Matrix Representation with Transposed Convolutions), our novel generalized tree representation for optimal structural search. Tart represents a tree model with a series of transposed convolutions that boost the speed of inference by avoiding the creation of transition matrices. As a result, Tart allows one to search for the best tree structure with a few design parameters, achieving higher classification accuracy than those of baseline models in feature-based datasets.
Language
eng
URI
https://aurora.ajou.ac.kr/handle/2018.oak/36857
https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85131320724&origin=inward
DOI
https://doi.org/2-s2.0-85131320724
Type
Conference
Funding
Publication of this article has been funded by the Basic Science Research Program through the National Research Foundation of Korea (2018R1A5A1060031).
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Lee, Sael Image
Lee, Sael이슬
Department of Software and Computer Engineering
Read More

Total Views & Downloads

File Download

  • There are no files associated with this item.