The paper derives a tripartite taxonomy of active learning phases (data-driven, transition, model-driven) from the structural unavoidability of shifts in dominant generalization mechanisms.
In: Pacific-Asia Conf
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Mechanism-Driven Theory of Phase Transitions in Active Learning
The paper derives a tripartite taxonomy of active learning phases (data-driven, transition, model-driven) from the structural unavoidability of shifts in dominant generalization mechanisms.