Research map / Architectures and efficiency / RNNs and normalization
Drop Clause: Enhancing Performance, Robustness and Pattern Recognition Capabilities of the Tsetlin Machine
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Add Drop Clause with p = .75 (ablation) | TM · IMDb | accuracy: 87.2 -> 91.27 | Helped |
| Add Drop Clause with p = .75 (ablation) | TM · R8 | accuracy: 96.16 -> 98.94 | Helped |
| Add Drop Clause with p = .75 (ablation) | TM · R52 | accuracy: 89.14 -> 92.75 | Helped |
| Add Drop Clause with p = .75 (ablation) | TM · MR | accuracy: 75.14 -> 78.67 | Helped |
| Use Drop Clause with p = .75 instead of p = .5 (ablation) | TM · TREC-6 | accuracy: 90.5 -> 89.9 | Hurt |
| Add Drop Clause with p = 0.25 (ablation) | CTM · MNIST | accuracy: 99.3 -> 99.45 | Helped |
| Use Drop Clause with p = 0.75 instead of p = 0.25 (ablation) | CTM · MNIST | accuracy: 99.45 -> 98.2 | Hurt |
| Add Drop Clause with p = 0.25 (ablation) | CTM · Fashion-MNIST | accuracy: 91.5 -> 92.5 | Helped |
| Add Drop Clause with p = 0.5 (ablation) | CTM · CIFAR-10 | accuracy: 69.3 -> 75.1 | Helped |
| Add Drop Clause with p = 0.5 (ablation) | CTM · CIFAR-100 | accuracy: 35.5 -> 45.2 | Helped |
| Add Drop Clause during training (ablation) | TM · MR (perturb) | accuracy: 73.12 -> 77.12 | Helped |
| Add Drop Clause during training (ablation) | CTM · MNIST-C | accuracy: 88.12 -> 91.75 | Helped |
Closest papers by meaning
- Generalized Convergence Analysis of Tsetlin Automaton Based Algorithms: A Probabilistic Approach to Concept Learning (AAAI 2025 · similarity 0.84)
- MIR-Bench: Can Your LLM Recognize Complicated Patterns via Many-Shot In-Context Reasoning? (NeurIPS 2025 · similarity 0.83)
- Remarkable Robustness of LLMs: Stages of Inference? (NeurIPS 2025 · similarity 0.82)
- Sample Smart, Not Hard: Correctness-First Decoding for Better Reasoning in LLMs (ICLR 2026 · similarity 0.82)
- Truth is Universal: Robust Detection of Lies in LLMs (NeurIPS 2024 · similarity 0.82)
- The Truth is in There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction (ICLR 2024 · similarity 0.82)
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.