AWhat changed
Introduced a homogeneous multi-branch residual block and cardinality as a model-design axis, with an equivalent grouped-convolution implementation for the selected bottleneck structure.
Scientific scope
Introduced a homogeneous multi-branch residual block and cardinality as a model-design axis, with an equivalent grouped-convolution implementation for the selected bottleneck structure. Grouped convolution, residual learning and split-transform-merge precedents are inherited. The displayed equivalence depends on the chosen homogeneous blocks and need not hold for arbitrary heterogeneous transformations. Matched-complexity accuracy advantages are empirical, not a theorem that cardinality always beats width/depth.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Grouped convolution, residual learning and split-transform-merge precedents are inherited. The displayed equivalence depends on the chosen homogeneous blocks and need not hold for arbitrary heterogeneous transformations. Matched-complexity accuracy advantages are empirical, not a theorem that cardinality always beats width/depth.
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.
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