Curse of Dimensionality in Interpretability

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Interpretability Illusionconcept0.759
  Cases where subspace interventions change model behaviour through parallel pathways rather than the target feature
• interpretabilityconcept0.749
  The capability to explain model predictions; a central theme of the paper, with disruption profiles as vehicle.
• "For interpretability, I don't think we even have the right definitions."quote0.737
  Ian Goodfellow quote used to illustrate the pre-paradigmatic state of interpretability research
• Multi-dimensional linear and non-linear interpretability methods have not been benchmarked on CausalGymquestion0.733
  Identified gap in benchmark coverage; only 1D linear methods are benchmarked
• Bottom-up interpretabilityconcept0.731
  An interpretability paradigm that explains computation in the model's own terms, rather than imposing top-down abstractions; VPD aims to realize this.
• Circuit Interpretabilityconcept0.729
  Advantage of DiffLogic CA over NCA — learned rules are pure binary logic circuits that can be visualized and analyzed
• Interpretability as Natural Scienceframework0.729
  Proposed paradigm for evaluating interpretability work through empirical falsifiability rather than benchmarks or user studies
• Neural Network Interpretabilityconcept0.725
  The field aimed at understanding what neural networks have learned; characterized as pre-paradigmatic in this paper