Abstract
Tsetlin Machine (TM) is a recent automaton-based algorithm for reinforcement learning. It has demonstrated competitive accuracy on many popular benchmarks while providing a natural interpretability. Due to its logical underpinning, it is amenable to hardware implementation with faster performance and higher energy efficiency than conventional Artificial Neural Networks. This paper introduces a multi-layer architecture of Tsetlin Machines with the aim to further boost TM performance via adoption of a hierarchical feature learning approach. This is seen as a way of creating hierarchical logic expressions from original Boolean literals, surpassing single-layer TMs in their ability to capture more complex patterns and high-level features. In this work we demonstrate that multi-layer TM considerably overperforms the single-layer TM architecture on several benchmarks while maintaining the ability to interpret its logic inference. However, it has also been shown that uncontrolled growth in the number of layers leads to overfitting.
More Information
Divisions: | School of Built Environment, Engineering and Computing |
---|---|
Identification Number: | https://doi.org/10.1109/ISTM62799.2024.10931518 |
Status: | Published |
Refereed: | Yes |
Publisher: | IEEE Xplore |
Additional Information: | © 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works |
SWORD Depositor: | Symplectic |
Depositing User (symplectic) | Deposited by Gorbenko, Anatoliy |
Date Deposited: | 22 Jul 2024 08:28 |
Last Modified: | 22 Jul 2025 12:46 |
Event Title: | 2024 International Symposium on the Tsetlin Machine (ISTM) |
Event Dates: | 29-30 Aug 2024 |
Item Type: | Conference or Workshop Item (Paper) |
Download
Note: this is the author's final manuscript and may differ from the published version which should be used for citation purposes.
| Preview
Export Citation
Explore Further
Read more research from the author(s):