A regularity tangent vector, computable with constant training overhead, turns the loss gradient at a candidate point into an estimate of its influence on model complexity for active learning.
Optimizing Millions of Hyperparameters by Implicit Differentiation, 2020
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Influence functions and regularity tangents for efficient active learning
A regularity tangent vector, computable with constant training overhead, turns the loss gradient at a candidate point into an estimate of its influence on model complexity for active learning.