The reviewed record of science sign in
Pith

arxiv: 2205.11716 · v2 · pith:IOODR5OT · submitted 2022-05-24 · cs.LG · math.PR· stat.ML

Randomly Initialized One-Layer Neural Networks Make Data Linearly Separable

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:IOODR5OTrecord.jsonopen to challenge →

classification cs.LG math.PRstat.ML
keywords neuralnetworkssetsinitializedlinearlyrandomlyseparablebound
0
0 comments X
read the original abstract

Recently, neural networks have demonstrated remarkable capabilities in mapping two arbitrary sets to two linearly separable sets. The prospect of achieving this with randomly initialized neural networks is particularly appealing due to the computational efficiency compared to fully trained networks. This paper contributes by establishing that, given sufficient width, a randomly initialized one-layer neural network can, with high probability, transform two sets into two linearly separable sets without any training. Moreover, we furnish precise bounds on the necessary width of the neural network for this phenomenon to occur. Our initial bound exhibits exponential dependence on the input dimension while maintaining polynomial dependence on all other parameters. In contrast, our second bound is independent of input dimension, effectively surmounting the curse of dimensionality. The main tools used in our proof heavily relies on a fusion of geometric principles and concentration of random matrices.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.