# Global Convergence of Frank Wolfe on One Hidden Layer Networks

arXiv: https://arxiv.org/abs/2002.02208 · HTML: https://mlsanity.com/paper/2002.02208/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Feb 2020.

## Built on by (most cited first, 1 of 1)

- [HardCoRe-NAS: Hard Constrained diffeRentiable Neural Architecture Search](https://mlsanity.com/paper/2102.11646/) (2021)

## Abstract (arXiv)

We derive global convergence bounds for the Frank Wolfe algorithm when training one hidden layer neural networks. When using the ReLU activation function, and under tractable preconditioning assumptions on the sample data set, the linear minimization oracle used to incrementally form the solution can be solved explicitly as a second order cone program. The classical Frank Wolfe algorithm then converges with rate O(1/T) where T is both the number of neurons and the number of calls to the oracle.