Toy model

The dataset used in the paper is a toy model, an MLP with a single linear hidden unit (1MLP) f (x) = w2w1x, which allows us to compare BP and PC exactly.

Data and Resources

Cite this as

Francesco Innocenti, Ryan Singh, Christopher L. Buckley (2024). Dataset: Toy model. https://doi.org/10.57702/8adwx07e

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.48550/arXiv.2305.18188
Author Francesco Innocenti
More Authors
Ryan Singh
Christopher L. Buckley