We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Hard-ODT: Hardware-Friendly Online Decision Tree Learning Algorithm and System

Abstract: Decision trees are machine learning models commonly used in various application scenarios. In the era of big data, traditional decision tree induction algorithms are not suitable for learning large-scale datasets due to their stringent data storage requirement. Online decision tree learning algorithms have been devised to tackle this problem by concurrently training with incoming samples and providing inference results. However, even the most up-to-date online tree learning algorithms still suffer from either high memory usage or high computational intensity with dependency and long latency, making them challenging to implement in hardware. To overcome these difficulties, we introduce a new quantile-based algorithm to improve the induction of the Hoeffding tree, one of the state-of-the-art online learning models. The proposed algorithm is light-weight in terms of both memory and computational demand, while still maintaining high generalization ability. A series of optimization techniques dedicated to the proposed algorithm have been investigated from the hardware perspective, including coarse-grained and fine-grained parallelism, dynamic and memory-based resource sharing, pipelining with data forwarding. Following this, we present Hard-ODT, a high-performance, hardware-efficient and scalable online decision tree learning system on a field-programmable gate array (FPGA) with system-level optimization techniques. Performance and resource utilization are modeled for the complete learning system for early and fast analysis of the trade-off between various design metrics. Finally, we propose a design flow in which the proposed learning system is applied to FPGA run-time power monitoring as a case study.
Comments: arXiv admin note: substantial text overlap with arXiv:2009.01431
Subjects: Machine Learning (cs.LG); Hardware Architecture (cs.AR)
DOI: 10.1109/TCAD.2020.3043328
Cite as: arXiv:2012.06272 [cs.LG]
  (or arXiv:2012.06272v1 [cs.LG] for this version)

Submission history

From: Zhe Lin [view email]
[v1] Fri, 11 Dec 2020 12:06:44 GMT (2357kb,D)

Link back to: arXiv, form interface, contact.