Ranking the information content of distance measures

Abstract

Real-world data typically contain a large number of features that are often heterogeneous in nature, relevance, and also units of measure. When assessing the similarity between data points, one can build various distance measures using subsets of these features. Using the fewest features but still retaining sufficient information about the system is crucial in many statistical learning approaches, particularly when data are sparse. We introduce a statistical test that can assess the relative information retained when using two different distance measures, and determine if they are equivalent, independent, or if one is more informative than the other. This in turn allows finding the most informative distance measure out of a pool of candidates. The approach is applied to find the most relevant policy variables for controlling the Covid-19 epidemic and to find compact yet informative representations of atomic structures, but its potential applications are wide ranging in many branches of science.

References in corpus (3)

• Machine Learning Unifies the Modelling of Materials and Molecules
• Machine-learning based interatomic potential for amorphous carbon
• DADApy: Distance-based Analysis of DAta-manifolds in Python

Cited by in corpus (21)

• DADApy: Distance-based Analysis of DAta-manifolds in Python
• Tensor-reduced atomic density representations
• Roadmap on machine learning glassy dynamics
• Do Machine-Learning Atomic Descriptors and Order Parameters Tell the Same Story? The Case of Liquid Water
• Beyond Local Structures In Critical Supercooled Water Through Unsupervised Learning
• Compact atomic descriptors enable accurate predictions via linear models
• ZundEig: The Structure of the Proton in Liquid Water From Unsupervised Learning
• Automatic feature selection and weighting in molecular systems using Differentiable Information Imbalance
• Robust inference of causality in high-dimensional dynamical processes from the Information Imbalance of distance ranks
• Data-driven discovery of statistically relevant information in quantum simulators
• Selecting Relevant Structural Features for Glassy Dynamics by Information Imbalance
• Aqueous Solution Chemistry In Silico and the Role of Data Driven Approaches
• A neural network potential with self-trained atomic fingerprints: a test with the mW water potential
• Maximum Information Extraction Via Clustering and Minimization of Shannon Entropy
• Linear scaling causal discovery from high-dimensional time series by dynamical community detection
• Towards a robust approach to infer causality in molecular systems satisfying detailed balance
• Detecting extreme event-driven causality
• Unsupervised detection of semantic correlations in big data
• Comparing the latent features of universal machine-learning interatomic potentials
• Compressing local atomic neighbourhood descriptors
• Assessing zero-shot generalisation behaviour in graph-neural-network interatomic potentials