Research map / Architectures and efficiency / RNNs and normalization
Massively Parallel and Asynchronous Tsetlin Machine Architecture Supporting Almost Constant-Time Scaling
Kuruge Darshana Abeyrathna, Bimal Bhattarai, Morten Goodwin, Saeed Rahimi Gorji, Ole-Christoffer Granmo, Lei Jiao, Rupsa Saha, Rohan K. Yadav
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use asynchronous clause threads with local voting tallies instead of synchronized sequential clause evaluation | Tsetlin Machine · classification datasets in Table 1 | accuracy | No clear effect |
| Use asynchronous clause threads with local voting tallies instead of indexed CPU learning | Tsetlin Machine · BBC Sports | speedup: 38.9$\times$ | Improved |
| Use asynchronous clause threads with local voting tallies instead of indexed CPU learning | Tsetlin Machine · 20 Newsgroup | speedup: 49.3$\times$ | Improved |
| Use asynchronous clause threads with local voting tallies instead of indexed CPU learning | Tsetlin Machine · IMDB | accuracy: 88.42 $\scriptstyle\pm$ 2.05 -> 89.92 $\scriptstyle\pm$ 0.23 | Improved |
| Increase the number of clauses | Regression Tsetlin Machine · Bike Sharing | MAE | Improved |
| Use CUDA parallelization rather than indexed CPU implementation | Regression Tsetlin Machine · Bike Sharing | speedup: 156.4$\times$ | Improved |
| Use CUDA parallelization rather than indexed CPU implementation | Regression Tsetlin Machine · BlogFeedback | speedup: 211.6$\times$ | Improved |
| Increase clause count from 20 to 2,560 | CUDA Tsetlin Machine · SEMEVAL | execution time: 40% | No clear effect |
| Use CUDA parallelization rather than indexed CPU implementation | Tsetlin Machine · JAVA | accuracy: 97.03 $\scriptstyle\pm$ 0.02 -> 97.53 $\scriptstyle\pm$ 0.02 | No clear effect |
| Use asynchronous clause threads with local voting tallies | Tsetlin Machine · Bike Sharing | MAE: 23.5$\scriptstyle\pm$0.04 -> 23.9$\scriptstyle\pm$0.08 | No clear effect |
Closest papers by meaning
- Think in Parallel, Answer as One: Logit Averaging for Open-Ended Reasoning (ICLR 2026 · similarity 0.82)
- Synergistic Tensor and Pipeline Parallelism (NeurIPS 2025 · similarity 0.82)
- Generalized Convergence Analysis of Tsetlin Automaton Based Algorithms: A Probabilistic Approach to Concept Learning (AAAI 2025 · similarity 0.82)
- ThreadWeaver: Adaptive Threading for Efficient Parallel Reasoning in Language Models (ICML 2026 · similarity 0.81)
- C-Voting: Confidence-Based Test-Time Voting without Explicit Energy Functions (ICLR 2026 · similarity 0.81)
- ATTS: Asynchronous Test-Time Scaling via Conformal Prediction (ICLR 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
