Not Just a Black Box: Learning Important Features Through Propagating Activation Differences

5 May 2016  ·  Avanti Shrikumar, Peyton Greenside, Anna Shcherbina, Anshul Kundaje ·

Note: This paper describes an older version of DeepLIFT. See for the newer version. Original abstract follows: The purported "black box" nature of neural networks is a barrier to adoption in applications where interpretability is essential. Here we present DeepLIFT (Learning Important FeaTures), an efficient and effective method for computing importance scores in a neural network. DeepLIFT compares the activation of each neuron to its 'reference activation' and assigns contribution scores according to the difference. We apply DeepLIFT to models trained on natural images and genomic data, and show significant advantages over gradient-based methods.

PDF Abstract


  Add Datasets introduced or used in this paper

Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.