We generate a local approximation or surrogate model that can be easily explained.
We present LEMON, an improvement to a popular explanation technique called LIME.
The difference between LIME and LEMON is in the way the synthetic data is sampled:
LIME samples points in the entire feature space, then reweights them according to the proximity to point to be explained.
Hence, LIME only finds few points (3) the area of interest.
We suggest instead to define the area of interest first, and then sample directly from the desired distribution.
As a result, we have much more data (15) to train a surrogate model on
The improved sampling approach in LEMON yields much more faithful explanations! More details can be found in the paper.
You can try out the LEMON technique for yourself by checking out the code on GitHub.Check GitHub
If you want to refer to our explanation technique, please cite our paper using the following BibTeX entry: