lit review resources

[singha53]

1. Interpretable bilinear attention network with domain adaptation improves drug–target prediction
2. Beyond explaining: Opportunities and challenges of XAI-based model improvement
3. Interpretable deep learning model to predict the molecular classification of endometrial cancer from haematoxylin and eosin-stained whole-slide images: a combined analysis of the PORTEC randomised trials and clinical cohorts
4. from linkedin post
   Majority of the machine learning models we use and develop are complex and not inherently interpretable. We can use explainability to understand how our models come up with their predictions, which is critical in most applications like healthcare and #drugdiscovery.

-SHAP (SHapley Additive exPlanations)
Library name for installing and importing: shap
URL: https://lnkd.in/ghYwsCMw

-Local Interpretable Model-Agnostic Explanations
Library name for installing and importing: lime
URL: https://lnkd.in/g3keAXun

-Shapash
Library name for installing and importing: shapash
URL: https://lnkd.in/g2Trfpdn

-ELI5
Library name for installing and importing: eli5
URL: https://lnkd.in/gjY5DKca

-Explainer dashboard
Library name for installing and importing: explainerdashboard
URL: https://lnkd.in/gPaGi99Z

-Dalex
Library name for installing and importing: dalex
URL: https://lnkd.in/g4U9yd5a

-OmniXAI
Library name for installing and importing: omnixai
URL: https://lnkd.in/gmRkgAP2

#pytorch specific library
-Captum
Library name for installing and importing: captum
URL: https://captum.ai/

There are other libraries specifically designed for bias and fairness that I will talk about in another post in near
future.

[singha53]

https://christophm.github.io/interpretable-ml-book/

[singha53]

https://sites.google.com/cs.washington.edu/csep590b