Chen/etal/2022a: Efficient Realization of Decision Trees for Real-Time Inference

Bibtype Article
Bibkey Chen/etal/2022a
Author Kuan-Hsun Chen and Chia-Hui Hsu and Christian Hakert and Sebastian Buschjäger and Chao-Lin Lee and Jenq-Kuen Lee and Katharina Morik and Jian-Jia Chen
Ls8autor Buschjäger, Sebastian
Morik, Katharina
Title Efficient Realization of Decision Trees for Real-Time Inference
Journal ACM Transactions on Embedded Computing Systems
Abstract For timing-sensitive edge applications, the demand for efficient lightweight machine learning solutions has increased recently. Tree ensembles are among the state-of-the-art in many machine learning applications. While single decision trees are comparably small, an ensemble of trees can have a significant memory footprint leading to cache locality issues, which are crucial to performance in terms of execution time. In this work, we analyze memory-locality issues of the two most common realizations of decision trees, i.e. native and if-else trees. We highlight, that both realizations demand a more careful memory layout to improve caching behavior and maximize performance. We adopt a probabilistic model of decision tree inference to find the best memory layout for each tree at the application layer. Further, we present an efficient heuristic to take architecture-dependent information into account thereby optimizing the given ensemble for a target computer architecture. Our code-generation framework, which is freely available on an open-source repository, produces optimized code sessions while preserving the structure and accuracy of the trees. With several real-world data sets, we evaluate the elapsed time of various tree realizations on server hardware as well as embedded systems for Intel and ARM processors. Our optimized memory layout achieves a reduction in execution time up to 75 % execution for server-class systems, and up to 70 % for embedded systems, respectively.
Year 2022
Projekt SFB876-A1
Issn 1539-9087

  • Privacy Policy
  • Imprint