K-ABENA combines threshold-based sample exclusion with Horvitz-Thompson inverse-probability reweighting to achieve unbiased gradient estimation while saving 28-54% of per-epoch backward-pass compute.
Curriculum learn- ing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
K-ABENA: K-Adaptive Backpropagation with Error-based N-exclusion Algorithm : (Compensated Loss-Based Sample Exclusion with Unbiased Gradient Estimation)
K-ABENA combines threshold-based sample exclusion with Horvitz-Thompson inverse-probability reweighting to achieve unbiased gradient estimation while saving 28-54% of per-epoch backward-pass compute.