FADE adapts to sequential covariate shift by combining a Fisher-weighted KL shift signal with a smoothed Fisher information regularizer, reporting gains on vision, text, and tabular benchmarks.
Sample selection bias correction theory
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Adapting to Fragmented and Evolving Data: A Fisher Information Perspective
FADE adapts to sequential covariate shift by combining a Fisher-weighted KL shift signal with a smoothed Fisher information regularizer, reporting gains on vision, text, and tabular benchmarks.