Bootstrap on convexified neural networks delivers theoretically consistent uncertainty quantification for CNNs with reduced computation via warm starts and transfer learning.
Title resolution pending
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
-
Uncertainty Quantification in CNN Through the Bootstrap of Convex Neural Networks
Bootstrap on convexified neural networks delivers theoretically consistent uncertainty quantification for CNNs with reduced computation via warm starts and transfer learning.