Block Kaczmarz methods with Tikhonov regularization eliminate recency bias in online preference learning, delivering higher alignment and inter-session stability than normalized baselines in simulations.
arXiv:2501.11673 (2025)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From Recency Bias to Stable Convergence Block Kaczmarz Methods for Online Preference Learning in Matchmaking Applications
Block Kaczmarz methods with Tikhonov regularization eliminate recency bias in online preference learning, delivering higher alignment and inter-session stability than normalized baselines in simulations.