This is a theoretical analysis of statistical complexity bounds for sparse neural networks with no empirical validation, experimental data, or clinical application.
Reported
Upper bound on Rademacher complex…$CWR\min\{k,\sqrt{sk/m}\log^{3/2}(2m)\}+kB/…
Complexity for zero-bias sparse n…$O(kWR/\sqrt m)$
Agnostic minimax excess-risk bound$\min\{1,\sqrt{s/(km)}\}$ up to logarithms