Research map / Learning theory and algorithms / Prediction and decision losses
Text Classification with Born's Rule
Emanuele Guidotti, Alfio Ferrara
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| add regularization (h=1) to classical configuration (ablation) | classical a=1,b=0,h=0 · 20Newsgroup | accuracy (%): 83.3 -> 82.0 | No clear effect |
| use BC weights (a=1/2,b=1,h=1) instead of classical configuration (a=1,b=0,h=0) (ablation) | Born Classifier · 20Newsgroup | accuracy (%): 83.3 -> 87.3 | Improved |
| select hyper-parameters by cross-validation instead of fixed prior (ablation) | Born Classifier · 20Newsgroup | accuracy (%): 87.3 -> 87.3 | No clear effect |
| use BC instead of multinomial naive Bayes | BC · 20Newsgroup | accuracy (%): 84.1 -> 87.3 | Improved |
| use BC instead of tuned baselines, runtime | BC · 20Newsgroup | runtime (s): 15.320 -> 0.043 | Improved |
| train Born Layer for 100 epochs instead of 1 epoch (ablation) | BL · 20NG | accuracy (%): 84.6 -> 87.1 | Improved |
| initialize Born Layer with BC weights (BC+BL, 1 epoch) instead of random (ablation) | BL · 20NG | accuracy (%): 84.6 -> 86.9 | Improved |
| initialize Born Layer with BC weights (10 epochs) (ablation) | BL · R52 | accuracy (%): 92.6 -> 95.2 | Improved |
| train BC+BL for 100 epochs instead of 10 (ablation) | BC+BL · R52 | accuracy (%): 95.2 -> 94.4 | Hurt |
| use BC instead of TextGCN | BC · R52 | accuracy (%): 93.6 -> 88.0 | Hurt |
| use BC instead of TextGCN | BC · R8 | accuracy (%): 97.1 -> 95.4 | Hurt |
| use BC instead of TextGCN, runtime | BC · 20NG | runtime (s): 1206.372 -> 0.001 | Improved |
Closest papers by meaning
- Quantum-Inspired Representation for Long-Tail Senses of Word Sense Disambiguation (AAAI 2023 · similarity 0.84)
- Quantum-Inspired Neural Network with Runge-Kutta Method (AAAI 2024 · similarity 0.82)
- Sublinear quantum algorithms for training linear and kernel-based classifiers (ICML 2019 · similarity 0.81)
- Quantum Interference Model for Semantic Biases of Glosses in Word Sense Disambiguation (AAAI 2024 · similarity 0.80)
- Quantum-inspired Non-homologous Representation Constraint Mechanism for Long-tail Senses of Word Sense Disambiguation (AAAI 2025 · similarity 0.80)
- Scaling Laws for Gradient Descent and Sign Descent for Linear Bigram Models under Zipf’s Law (NeurIPS 2025 · similarity 0.79)
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.
