Hyperblock simplification (R2A, R2B, disjunctive units) cuts model complexity by 92-99% while holding accuracy within 1-2% of prior hyperblock and black-box baselines on WBC and MNIST.
Discovering interpretable machine learning models in parallel coordinates,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Fully Explainable Classification Models Using Hyperblocks
Hyperblock simplification (R2A, R2B, disjunctive units) cuts model complexity by 92-99% while holding accuracy within 1-2% of prior hyperblock and black-box baselines on WBC and MNIST.