Letter #21 - In which we haven’t decided anything yet.
The only thing worse than being incompetent, or being unkind, or being evil, is being indecisive.
Come a little gamma ray
Standing in a hurricane
Your brains are bored
Like a refugee
From the houses burning
And the heat wave’s
Calling your name
Many machine learning systems look at some kind of complicated input (say, an image) and produce a simple output (a label like, “cat”). By contrast, the goal of a generative model is something like the opposite: take a small piece of input—perhaps a few random numbers—and produce a complex output, like an image of a realistic-looking face. A generative adversarial network (GAN) is an especially effective type of generative model, introduced only a few years ago, which has been a subject of intense interest in the machine learning community.
Machine learning models remain mostly black boxes. Understanding the reasons behind predictions is, however, quite important in assessing trust, which is fundamental if one plans to take action based on a prediction, or when choosing whether to deploy a new model. Such understanding also provides insights into the model, which can be used to transform an untrustworthy model or prediction into a trustworthy one.
Nice introduction to the LIME explaining tool for machine learning models. Impressive what it can do re: text classification.
Not sure what you just read? Take a look at this post.
Enjoyed it? Subscribe to my tiny newsletter and get new issues straight to your inbox. Every week, on Tuesdays.