Saved in:
Bibliographic Details
Main Authors: Zeng, Yefan, Duan, Shengyu, Shafik, Rishad, Yakovlev, Alex
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.15653
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918162616811520
author Zeng, Yefan
Duan, Shengyu
Shafik, Rishad
Yakovlev, Alex
author_facet Zeng, Yefan
Duan, Shengyu
Shafik, Rishad
Yakovlev, Alex
contents The Tsetlin Machine (TM) offers high-speed inference on resource-constrained devices such as CPUs. Its logic-driven operations naturally lend themselves to parallel execution on modern CPU architectures. Motivated by this, we propose an efficient software implementation of the TM by leveraging instruction-level bitwise operations for compact model representation and accelerated processing. To further improve inference speed, we introduce an early exit mechanism, which exploits the TM's AND-based clause evaluation to avoid unnecessary computations. Building upon this, we propose a literal Reorder strategy designed to maximize the likelihood of early exits. This strategy is applied during a post-training, pre-inference stage through statistical analysis of all literals and the corresponding actions of their associated Tsetlin Automata (TA), introducing negligible runtime overhead. Experimental results using the gem5 simulator with an ARM processor show that our optimized implementation reduces inference time by up to 96.71% compared to the conventional integer-based TM implementations while maintaining comparable code density.
format Preprint
id arxiv_https___arxiv_org_abs_2510_15653
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fast and Compact Tsetlin Machine Inference on CPUs Using Instruction-Level Optimization
Zeng, Yefan
Duan, Shengyu
Shafik, Rishad
Yakovlev, Alex
Machine Learning
The Tsetlin Machine (TM) offers high-speed inference on resource-constrained devices such as CPUs. Its logic-driven operations naturally lend themselves to parallel execution on modern CPU architectures. Motivated by this, we propose an efficient software implementation of the TM by leveraging instruction-level bitwise operations for compact model representation and accelerated processing. To further improve inference speed, we introduce an early exit mechanism, which exploits the TM's AND-based clause evaluation to avoid unnecessary computations. Building upon this, we propose a literal Reorder strategy designed to maximize the likelihood of early exits. This strategy is applied during a post-training, pre-inference stage through statistical analysis of all literals and the corresponding actions of their associated Tsetlin Automata (TA), introducing negligible runtime overhead. Experimental results using the gem5 simulator with an ARM processor show that our optimized implementation reduces inference time by up to 96.71% compared to the conventional integer-based TM implementations while maintaining comparable code density.
title Fast and Compact Tsetlin Machine Inference on CPUs Using Instruction-Level Optimization
topic Machine Learning
url https://arxiv.org/abs/2510.15653