ResearchOct 5, 2026 · 11:001 source

Approximation Property of Dropout Neural Networks: Sobolev Rates and Confidence Bounds

From arxiv.org

AI;DR

The study analyzed approximation of the unit ball of $W^{n,\infty}([0,1]^d)$ by ReLU networks whose edges are retained independently with probability $p$, constructing networks of constant depth and size $\widetilde O_{n,d}(p^{-9}\varepsilon^{-\max\{d/n,2\}} \log(1/\delta))$ that guarantee approximation error bounds with probability at least $1-\delta$.

Read the sourceSign in to suggest

Sources

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today