2026
CLM-AI-003613>= VCdim = Omega(W L log(W/L))LOWER BOUNDThe cited theorem establishes VCdim = Omega(W L log(W/L)) under the recorded model.Source: Nearly-tight VC-dimension and Pseudodimension Bounds for Piecewise Linear Neural Networks — ReLU-network VC-dimension lower bound ↗
PROVEN