A knowledge-guided framework produces a differentiable surrogate for Minkowski functionals on precipitation images via Lipschitz-constrained CNNs, validated on radar data but revealing a stability-versus-detail trade-off in super-resolution.
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
-
Emulating Non-Differentiable Metrics via Knowledge-Guided Learning: Introducing the Minkowski Image Loss
A knowledge-guided framework produces a differentiable surrogate for Minkowski functionals on precipitation images via Lipschitz-constrained CNNs, validated on radar data but revealing a stability-versus-detail trade-off in super-resolution.