Network In Network: local micro-networks and global average pooling

What changed

Replaced local linear filters by shared nonlinear micro-MLPs and used global averaging of final feature maps for classification.

Scientific scope

Replaced local linear filters by shared nonlinear micro-MLPs and used global averaging of final feature maps for classification. Inherits CNNs/MLPs and pooling; specific contribution is their local micro-network/global-readout design, not invention of every1×1convolution or all averaging. Feature-map confidence/interpretability language is architectural motivation, not calibrated semantic probabilities or universal interpretability.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Inherits CNNs/MLPs and pooling; specific contribution is their local micro-network/global-readout design, not invention of every1×1convolution or all averaging. Feature-map confidence/interpretability language is architectural motivation, not calibrated semantic probabilities or universal interpretability.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• Network In Network

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction