arxivst stuff from arxiv that you should probably bookmark

Learning Important Features Through Propagating Activation Differences

Abstract · Apr 10, 2017 02:23 ·

deeplift reference importance gradient neurons output greenside cs-cv cs-lg cs-ne

Arxiv Abstract

  • Avanti Shrikumar
  • Peyton Greenside
  • Anshul Kundaje

The purported “black box”’ nature of neural networks is a barrier to adoption in applications where interpretability is essential. Here we present DeepLIFT (Deep Learning Important FeaTures), a method for decomposing the output prediction of a neural network on a specific input by backpropagating the contributions of all neurons in the network to every feature of the input. DeepLIFT compares the activation of each neuron to its ‘reference activation’ and assigns contribution scores according to the difference. By optionally giving separate consideration to positive and negative contributions, DeepLIFT can also reveal dependencies which are missed by other approaches. Scores can be computed efficiently in a single backward pass. We apply DeepLIFT to models trained on MNIST and simulated genomic data, and show significant advantages over gradient-based methods. A detailed video tutorial on the method is at and code is at

Read the paper (pdf) »