npj Digital Medicine (May 2023)

Theory and rationale of interpretable all-in-one pattern discovery and disentanglement system

  • Andrew K. C. Wong,
  • Pei-Yuan Zhou,
  • Annie E.-S. Lee

DOI
https://doi.org/10.1038/s41746-023-00816-9
Journal volume & issue
Vol. 6, no. 1
pp. 1 – 15

Abstract

Read online

Abstract In machine learning (ML), association patterns in the data, paths in decision trees, and weights between layers of the neural network are often entangled due to multiple underlying causes, thus masking the pattern-to-source relation, weakening prediction, and defying explanation. This paper presents a revolutionary ML paradigm: pattern discovery and disentanglement (PDD) that disentangles associations and provides an all-in-one knowledge system capable of (a) disentangling patterns to associate with distinct primary sources; (b) discovering rare/imbalanced groups, detecting anomalies and rectifying discrepancies to improve class association, pattern and entity clustering; and (c) organizing knowledge for statistically supported interpretability for causal exploration. Results from case studies have validated such capabilities. The explainable knowledge reveals pattern-source relations on entities, and underlying factors for causal inference, and clinical study and practice; thus, addressing the major concern of interpretability, trust, and reliability when applying ML to healthcare, which is a step towards closing the AI chasm.