Two new non-convex signal-lasso penalties are introduced for binary network reconstruction, but the product-penalty update is incorrect and the no-tuning claim is only asymptotically true.
The most of real networks exhibit sparsely connected properties, and the connection parameter is a signal connected (0 or 1)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.ST 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Signal Lasso with Non-Convex Penalties for Efficient Network Reconstruction and Topology Inference
Two new non-convex signal-lasso penalties are introduced for binary network reconstruction, but the product-penalty update is incorrect and the no-tuning claim is only asymptotically true.