Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

[Link] Introduction to Local Interpretable Model-Agnostic Explanations (LIME)

4 Post author: Gunnar_Zarncke 09 February 2017 08:29AM

Comments (2)

Comment author: Lumifer 09 February 2017 03:57:47PM 1 point [-]

It's an interesting idea, though I wouldn't call it "model-agnostic".

Basically they're jiggering the inputs and figuring out what you can't change without the prediction (classification) changing as well. In effect they are answering the question "given this model, which input values are essential to producing this particular output?"

Comment author: Gunnar_Zarncke 09 February 2017 08:30:16AM 0 points [-]

Another (earlier) blog post about it: https://homes.cs.washington.edu/~marcotcr/blog/lime/