DeepLIFT: Learning Important Features Through Propagating Activation Differences

DeepLIFT is a method for assigning feature importance that compares a neuron's activation to its'reference', where the reference is the activation that the neuron has when the network is provided a'reference input'.

Data and Resources

Cite this as

Avanti Shrikumar, Peyton Greenside, Anna Y. Shcherbina, Anshul Kundaje (2024). Dataset: DeepLIFT: Learning Important Features Through Propagating Activation Differences. https://doi.org/10.57702/ns309vo8

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.48550/arXiv.1605.01713
Author Avanti Shrikumar
More Authors
Peyton Greenside
Anna Y. Shcherbina
Anshul Kundaje
Homepage https://arxiv.org/abs/1704.02685