A Novel Multi-step Finite-State Automaton for Arbitrarily Deterministic Tsetlin Machine Learning

Abeyrathna, K. Darshana; Shafik, Rishad; Yakovlev, Alex; Wheeldon, Alan; Lei, Jie; Goodwin, Morten

doi:10.1007/978-3-030-63799-6_8

Cited by 7 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…В [3] авторы описывают новый конечный обучающий автомат, который был разработан для снижения энергозатрат при преобразовании автоматов в детерминированные. Новый алгоритм содержит особый параметр для нахождения компромисса между точностью и потреблением энергии, что позволяет упростить машинное обучение.…”

Section: анализ предметной областиunclassified

Application of specialized DFA for time series forecasting

Pavlova

Rakitskiy

2022

jour

View full text Add to dashboard Cite

In this paper the possibility of using deterministic finite automaton to forecast time series in real time is considered. To solve this problem, a modification of the automaton with 10 reading heads designed to recognize multilinear sequences is proposed. The article presents modifications of this automaton, algorithms for their implementation, and demonstrates the results for various time series. In addition, the paper presents a method for changing the algorithm of the automaton ensuring its step-by-step execution.

show abstract

Section: анализ предметной областиunclassified

Application of specialized DFA for time series forecasting

Pavlova

Rakitskiy

2022

jour

View full text Add to dashboard Cite

show abstract

“…Exploiting the natural logic underpinning there are currently ongoing efforts in establishing explainability of TMs [17]. Deterministic implementation of clause selection in TM, reported by [49], is a promising direction to this end.…”

Section: The Tsetlin Machinementioning

confidence: 99%

Low-Power Audio Keyword Spotting Using Tsetlin Machines

Lei

Rahman

Shafik

et al. 2021

JLPEA

Self Cite

View full text Add to dashboard Cite

The emergence of artificial intelligence (AI) driven keyword spotting (KWS) technologies has revolutionized human to machine interaction. Yet, the challenge of end-to-end energy efficiency, memory footprint and system complexity of current neural network (NN) powered AI-KWS pipelines has remained ever present. This paper evaluates KWS utilizing a learning automata powered machine learning algorithm called the Tsetlin Machine (TM). Through significant reduction in parameter requirements and choosing logic over arithmetic-based processing, the TM offers new opportunities for low-power KWS while maintaining high learning efficacy. In this paper, we explore a TM-based keyword spotting (KWS) pipeline to demonstrate low complexity with faster rate of convergence compared to NNs. Further, we investigate the scalability with increasing keywords and explore the potential for enabling low-power on-chip KWS.

show abstract

“…Each pattern is formed by ANDing selected entries x k in the input vector x, or their negations xk . 3 The pattern…”

Section: A14 Pattern Constructionmentioning

confidence: 99%

Coalesced Multi-Output Tsetlin Machines with Clause Sharing

Glimsdal,

Granmo

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Using finite-state machines to learn patterns, Tsetlin machines (TMs) have obtained competitive accuracy and learning speed across several benchmarks, with frugal memoryand energy footprint. A TM represents patterns as conjunctive clauses in propositional logic (AND-rules), each clause voting for or against a particular output. While efficient for single-output problems, one needs a separate TM per output for multi-output problems. Employing multiple TMs hinders pattern reuse because each TM then operates in a silo. In this paper, we introduce clause sharing, merging multiple TMs into a single one. Each clause is related to each output by using a weight. A positive weight makes the clause vote for output 1, while a negative weight makes the clause vote for output 0. The clauses thus coalesce to produce multiple outputs. The resulting coalesced Tsetlin machine (CoTM) simultaneously learns both the weights and the composition of each clause by employing interacting Stochastic Searching on the Line (SSL) and Tsetlin automata (TAs) teams. Our empirical results on MNIST, Fashion-MNIST, and Kuzushiji-MNIST show that CoTM obtains significantly higher accuracy than TM on 50-to 1K-clause configurations, indicating an ability to repurpose clauses. E.g., accuracy goes from 71.99% to 89.66% on Fashion-MNIST when employing 50 clauses per class (22 Kb memory). While TM and CoTM accuracy is similar when using more than 1K clauses per class, CoTM reaches peak accuracy 3ˆfaster on MNIST with 8K clauses. We further investigate robustness towards imbalanced training data. Our evaluations on imbalanced versions of IMDb-and CIFAR10 data show that CoTM is robust towards high degrees of class imbalance. Being able to share clauses, we believe CoTM will enable new TM application domains that involve multiple outputs, such as learning language models and auto-encoding.

show abstract

A Novel Multi-step Finite-State Automaton for Arbitrarily Deterministic Tsetlin Machine Learning

Cited by 7 publications

References 15 publications

Application of specialized DFA for time series forecasting

Application of specialized DFA for time series forecasting

Low-Power Audio Keyword Spotting Using Tsetlin Machines

Coalesced Multi-Output Tsetlin Machines with Clause Sharing

Contact Info

Product

Resources

About