Controlling Confusion via Generalisation BoundsDownload PDF

16 May 2022 (modified: 03 Jul 2024)NeurIPS 2022 SubmittedReaders: Everyone
Keywords: PAC-Bayes, Generalisation bounds, Multiclass classification
Abstract: We establish new generalisation bounds for multiclass classification by abstracting to a more general setting of discretised error types. Extending the PAC-Bayes theory, we are hence able to provide fine-grained bounds on performance for multiclass classification, as well as applications to other learning problems including discretisation of regression losses. Tractable training objectives are derived from the bounds. The bounds are uniform over all weightings of the discretised error types and thus can be used to bound weightings not foreseen at training, including the full confusion matrix in the multiclass classification case.
TL;DR: A new type of generalisation bound providing more informative measures of performance.
Supplementary Material: pdf
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/controlling-confusion-via-generalisation/code)
16 Replies

Loading