Skip to content

Implement NTK scorer  #274

@mdbenito

Description

@mdbenito

Introduced in Zhaoxuan Wu, Yao Shu, and Bryan Kian Hsiang Low, “DAVINZ: Data Valuation Using Deep Neural Networks at Initialization,” in Proceedings of the 39th International Conference on Machine Learning (International Conference on Machine Learning, PMLR, 2022), 24150–76.

It's a scorer using the Neural Tangent Kernel on untrained networks with a correction to estimate generalisation error. Implementation available in the supp. materials of the paper.

(Maybe relevant, JAX library for NTK: https://github.com/google/neural-tangents, and paper about fast finite-width NTK computation)

Metadata

Metadata

Assignees

No one assigned

    Labels

    new-methodImplementation of new algorithms for valuation or influence functions

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions