NANAAug 13, 2018

New error bounds for deep networks using sparse grids

arXiv:1712.08688120 citationsh-index: 13
AI Analysis

Provides theoretical guarantees for deep network approximation, relevant to researchers in approximation theory and deep learning theory.

The paper proves new error bounds for deep ReLU networks approximating multivariate functions, showing that the curse of dimensionality can be mitigated by leveraging connections with sparse grids.

We prove a theorem concerning the approximation of multivariate functions by deep ReLU networks. We present new error estimates for which the curse of the dimensionality is lessened by establishing a connection with sparse grids.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes