Pith. sign in

REVIEW 12 cited by

The Tsetlin Machine -- A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1804.01508 v15 pith:DA44KYOH submitted 2018-04-04 cs.AI cs.CVcs.LG

Ole-Christoffer Granmo This is my paper · ORCID
classification cs.AIcs.CVcs.LG
keywords tsetlinmachinepropositionalrecognitionaccuracyformulasgamelearning
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Although simple individually, artificial neurons provide state-of-the-art performance when interconnected in deep networks. Arguably, the Tsetlin Automaton is an even simpler and more versatile learning mechanism, capable of solving the multi-armed bandit problem. Merely by means of a single integer as memory, it learns the optimal action in stochastic environments through increment and decrement operations. In this paper, we introduce the Tsetlin Machine, which solves complex pattern recognition problems with propositional formulas, composed by a collective of Tsetlin Automata. To eliminate the longstanding problem of vanishing signal-to-noise ratio, the Tsetlin Machine orchestrates the automata using a novel game. Further, both inputs, patterns, and outputs are expressed as bits, while recognition and learning rely on bit manipulation, simplifying computation. Our theoretical analysis establishes that the Nash equilibria of the game align with the propositional formulas that provide optimal pattern recognition accuracy. This translates to learning without local optima, only global ones. In five benchmarks, the Tsetlin Machine provides competitive accuracy compared with SVMs, Decision Trees, Random Forests, Naive Bayes Classifier, Logistic Regression, and Neural Networks. We further demonstrate how the propositional formulas facilitate interpretation. In conclusion, we believe the combination of high accuracy, interpretability, and computational simplicity makes the Tsetlin Machine a promising tool for a wide range of domains.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 12 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Autonomous Collaborative Learning Among an Ensemble of Tsetlin Machines with Consensus-Based Inference

    cs.LG 2026-07 conditional novelty 6.0 of 10

    A two-layer Tsetlin Machine ensemble with gossip-based vote sharing matches centralized accuracy on several benchmarks without exchanging raw data.

  2. Macroeconomic Message Passing for Anticipating Foreign Exchange Regime Changes: A Deep Logical Learning Approach using Graph Tsetlin Machines

    cs.CE 2026-07 conditional novelty 6.0 of 10

    A Graph Tsetlin Machine with hypervectorized macro multigraphs and message-passing clauses anticipates four USD/JPY regimes, reaching 70.7% overall OOS accuracy and beating reduced-graph and CoTM baselines on stagnant...

  3. Runtime Tunable Tsetlin Machines for Edge Inference on eFPGAs

    cs.AR 2025-02 conditional novelty 6.0 of 10

    A runtime-reconfigurable FPGA accelerator for compressed Tsetlin machines uses fewer logic resources than prior designs and reports up to 129x energy savings versus a microcontroller baseline.

  4. ETHEREAL: Energy-efficient and High-throughput Inference using Compressed Tsetlin Machine

    cs.LG 2025-02 conditional novelty 6.0 of 10

    A new training-time pruning method for Tsetlin machines removes literals shared by positive and negative clauses, cutting model size by up to 87.54% with at most 3.38% accuracy loss.

  5. An All-digital 8.6-nJ/Frame 65-nm Tsetlin Machine Image Classification Accelerator

    cs.LG 2025-01 conditional novelty 6.0 of 10

    A 65 nm all-digital Tsetlin machine accelerator chip classifies MNIST images at 8.6 nJ per frame and 97.42% accuracy, the lowest fully digital EPC reported for this benchmark.

  6. Omni TM-AE: A Scalable and Interpretable Embedding Model Using the Full Tsetlin Machine State Space

    cs.LG 2025-05 reject novelty 5.0 of 10

    A Tsetlin Machine autoencoder that averages signed automaton states over all literals produces single-phase, reusable word embeddings with results close to mainstream embedding models.

  7. IMPACT:InMemory ComPuting Architecture Based on Y-FlAsh Technology for Coalesced Tsetlin Machine Inference

    cs.AR 2024-12 conditional novelty 5.0 of 10

    A Y-Flash in-memory computing architecture for coalesced Tsetlin machine inference is presented, simulating 96.3 percent MNIST accuracy with improved energy efficiency.

  8. Leveraging Interpretable Tsetlin Machine for PDF Malware Detection

    cs.CR 2026-07 conditional novelty 4.5 of 10

    Tsetlin Machine detects PDF malware at 98.02% accuracy with clause-based interpretability on the RIT-PDFMal-2026 static-feature dataset.

  9. Explainable and Hardware-Efficient Jamming Detection for 5G Networks Using the Convolutional Tsetlin Machine

    eess.SP 2026-03 conditional novelty 4.0 of 10

    A Tsetlin-machine classifier on 5G SSB features achieves 91.5% jamming-detection accuracy with much faster training and a 14x smaller memory footprint than a CNN, at the cost of a 5-point accuracy gap and slower CPU i...

  10. A Survey of Explainable Reinforcement Learning: Targets, Methods and Needs

    cs.AI 2025-07 conditional novelty 4.0 of 10

    A survey of 250+ explainable-reinforcement-learning papers proposes a What/How taxonomy and reports that sequence-level explanations are rare (11 works) compared with policy-level (175) and action-level (89) ones.

  11. Uncertainty Quantification in the Tsetlin Machine

    cs.LG 2025-07 reject novelty 4.0 of 10

    A Tsetlin machine's class sum is mapped to a probability score by P=(1+v/T)/2 and used for uncertainty quantification, but the derivation is heuristic and the validation is weak.

  12. Adversarial Attacks on AI-Generated Text Detection Models: A Token Probability-Based Approach Using Embeddings

    cs.CL 2025-01 conditional novelty 4.0 of 10

    A hybrid synonym-and-embedding word substitution lowers Fast-DetectGPT AUROC to 0.2744 on XSum and 0.3532 on SQuAD, according to the paper.

Pith tools