Prototype Based Classification from Hierarchy to Fairness

Mycal Tucker, Julie A. Shah
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:21884-21900, 2022.

Abstract

Artificial neural nets can represent and classify many types of high-dimensional data but are often tailored to particular applications – e.g., for “fair” or “hierarchical” classification. Once an architecture has been selected, it is often difficult for humans to adjust models for a new task; for example, a hierarchical classifier cannot be easily transformed into a fair classifier that shields a protected field. Our contribution in this work is a new neural network architecture, the concept subspace network (CSN), which generalizes existing specialized classifiers to produce a unified model capable of learning a spectrum of multi-concept relationships. We demonstrate that CSNs reproduce state-of-the-art results in fair classification when enforcing concept independence, may be transformed into hierarchical classifiers, or may even reconcile fairness and hierarchy within a single classifier. The CSN is inspired by and matches the performance of existing prototype-based classifiers that promote interpretability.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-tucker22a, title = {Prototype Based Classification from Hierarchy to Fairness}, author = {Tucker, Mycal and Shah, Julie A.}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {21884--21900}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/tucker22a/tucker22a.pdf}, url = {https://proceedings.mlr.press/v162/tucker22a.html}, abstract = {Artificial neural nets can represent and classify many types of high-dimensional data but are often tailored to particular applications – e.g., for “fair” or “hierarchical” classification. Once an architecture has been selected, it is often difficult for humans to adjust models for a new task; for example, a hierarchical classifier cannot be easily transformed into a fair classifier that shields a protected field. Our contribution in this work is a new neural network architecture, the concept subspace network (CSN), which generalizes existing specialized classifiers to produce a unified model capable of learning a spectrum of multi-concept relationships. We demonstrate that CSNs reproduce state-of-the-art results in fair classification when enforcing concept independence, may be transformed into hierarchical classifiers, or may even reconcile fairness and hierarchy within a single classifier. The CSN is inspired by and matches the performance of existing prototype-based classifiers that promote interpretability.} }
Endnote
%0 Conference Paper %T Prototype Based Classification from Hierarchy to Fairness %A Mycal Tucker %A Julie A. Shah %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-tucker22a %I PMLR %P 21884--21900 %U https://proceedings.mlr.press/v162/tucker22a.html %V 162 %X Artificial neural nets can represent and classify many types of high-dimensional data but are often tailored to particular applications – e.g., for “fair” or “hierarchical” classification. Once an architecture has been selected, it is often difficult for humans to adjust models for a new task; for example, a hierarchical classifier cannot be easily transformed into a fair classifier that shields a protected field. Our contribution in this work is a new neural network architecture, the concept subspace network (CSN), which generalizes existing specialized classifiers to produce a unified model capable of learning a spectrum of multi-concept relationships. We demonstrate that CSNs reproduce state-of-the-art results in fair classification when enforcing concept independence, may be transformed into hierarchical classifiers, or may even reconcile fairness and hierarchy within a single classifier. The CSN is inspired by and matches the performance of existing prototype-based classifiers that promote interpretability.
APA
Tucker, M. & Shah, J.A.. (2022). Prototype Based Classification from Hierarchy to Fairness. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:21884-21900 Available from https://proceedings.mlr.press/v162/tucker22a.html.

Related Material