JavaScript isn't enabled in your browser, so this file can't be opened. Enable and reload.
Quiz for 15.11.22
What Does LIME Really See in Images?
Sign in to Google
to save your progress.
Learn more
* Indicates required question
Email
*
Your email
Pseudonym/Student_ID
*
Your answer
What kind of data can be served to LIME?
Tabular
Images
Audio
Text
What is required for the interpretable coefficients to converge?
The baseline is composed of superpixels containing their average color instead of just black
Small number of perturbed examples
The data is images
Large number of perturbed example
Clear selection
Of the following explainability techniques, which was LIME primarily compared to?
Integrated Gradients
Protodash
Global Interpretation via Recursive Partitioning (GIRP)
Counterfactual Instances
Tree Surrogates
Explainable Boosting Machine (EBM)
Contrastive Explanation Method (CEM)
SHAP
Accumulated Local Effects (ALE)
Scalable Bayesian Rule Lists
Anchors
Morris Sensitivity Analysis
Partial Dependence Plot
Permutation Importance
Clear selection
"The authors provide a method to compute the vector at which the interpretable coefficients given by LIME converge." - Is this correct and under what conditions?
Yes, but only for shape detectors
No, but they did manage to approximate it
No, they provided a proof that this is theoretically impossible (see appendix)
Yes, but only for linear models
Clear selection
Submit
Clear form
Never submit passwords through Google Forms.
This content is neither created nor endorsed by Google.
Report Abuse
-
Terms of Service
-
Privacy Policy
Forms
