paper
referenced-only
2017
paper:a-2017-interpretable

Interpretable explanations of black boxes by meaningful perturbation

ByA Vedaldi·RC Fong

Similar preprints — Semantic Scholar

Cited by (3)

  • Stochastic Parameter Decomposition

    Rank-one subcomponents trained against stochastic ablation masks recover ground-truth mechanisms in toy networks that Attribution-based Parameter Decomposition (APD) could not decompose, while removin

  • Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds

    Lifting the featurizer's atom from a direction to a block of directions recovers concept manifolds that SAEs shatter, and block sparsity is the MAP-matched prior for a representation that is a sparse

  • Zoom In: An Introduction to Circuits

    The Circuits framework proposes that neural network internals are legible at the level of individual neurons and their weighted connections, advancing three speculative claims: features (directions in