Prototype Based Classification from Hierarchy to Fairness
This addresses the need for flexible and interpretable classifiers in machine learning, though it appears incremental as it generalizes existing specialized methods.
The paper tackles the problem of neural networks being tailored to specific tasks like fair or hierarchical classification, making them difficult to adapt to new tasks, and introduces the concept subspace network (CSN) as a unified model that reproduces state-of-the-art results in fair classification and can be transformed into hierarchical classifiers.
Artificial neural nets can represent and classify many types of 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 even reconcile fairness and hierarchy within a single classifier. The CSN is inspired by existing prototype-based classifiers that promote interpretability.