Metric Space Magnitude for Evaluating the Diversity of Latent Representations

no code implementations • 27 Nov 2023 • Katharina Limbeck, Rayna Andreeva, Rik Sarkar, Bastian Rieck

We develop a family of magnitude-based measures of the intrinsic diversity of latent representations, formalising a novel notion of dissimilarity between magnitude functions of finite metric spaces.

Dimensionality Reduction Representation Learning

Metric Space Magnitude and Generalisation in Neural Networks

no code implementations • 9 May 2023 • Rayna Andreeva, Katharina Limbeck, Bastian Rieck, Rik Sarkar

Deep learning models have seen significant successes in numerous applications, but their inner workings remain elusive.