I see two different repos for "DeepExplain" neural network implementation. How do they differ? Are they the same algorithm but implemented by different people?

Deep learning example with DeepExplainer (TensorFlow/Keras models)


They are two implementations of different algorithms.

SHAP offers two model-specific explainer DeepShap and GradientShap for explaining neural network models. The former combine the idea of DeepLift and Shapley values, the latter combines the idea of IntegratedGradients and Shapley values. SHAP also offers a model-agnostic algorithm named KernalShap, which uses ideas of LIME but with modified weights for local perturbated samples.

The DeepExplain package offers different algorithms in a unified framework, including DeepLift and IntegratedGradients.

Here's a useful article of SHAP https://christophm.github.io/interpretable-ml-book/shap.html. Some brief explanations of these algorithms can be also found in Captum's doc, https://captum.ai/docs/algorithms.


It seems as both repositories are for 'network explainers'.

The SHAP repository looks like the official implementation of the algorithms they cite in their citation section.

While DeepExplain repo has a broader cohort of algorithms of method also implementation, which I assume are not the the official implementations.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.