LIME

William Vorhies discusses a new technical paper on Local Interpretable Model-Agnostic Explanations:

What the model actually used for classification were these: ‘posting’, ‘host’, ‘NNTP’, ‘EDU’, ‘have’, ‘there’.  These are meaningless artifacts that appear in both the training and test sets and have nothing to do with the topic except that, for example, the word “posting” (part of the email header) appears in 21.6% of the examples in the training set but only two times in the class “Christianity.”

Is this model going to generalize?  Absolutely not.

An Example from Image Processing

In this example using Google’s Inception NN on arbitrary images the objective was to correctly classify “tree frogs”.  The classifier was correct in about 54% of cases but also interpreted the image as a pool table (7%) and a balloon (5%).

Looks like an interesting paper.  Click through for a link to the paper.

Related Posts

Comparing TensorFlow Versus PyTorch

Anirudh Rao compares PyTorch to TensorFlow: For small-scale server-side deployments both frameworks are easy to wrap in e.g. a Flask web server. For mobile and embedded deployments, TensorFlow works really well. This is more than what can be said of most other deep learning frameworks including PyTorch. Deploying to Android or iOS does require a non-trivial amount of work in TensorFlow. You don’t have to rewrite the entire inference portion of your model in Java or C++. […]

Read More

Using R To Hit Azure ML From Power BI

Leila Etaati shows how you can use R to hit an Azure ML endpoint to populate a data set in Power BI: You need to create a model in Azure ML Studio and create a web service for it. The traditional example in Predict a passenger on Titanic ship is going to survived or not? […]

Read More

Categories

September 2016
MTWTFSS
« Aug Oct »
 1234
567891011
12131415161718
19202122232425
2627282930