Summer QoTD #12

It is widely accepted that modern deep learning systems do not have a deep understanding of the task they are trained to do, but rather use superficial correlations to arrive at answers. This is similar to a supposed German prodigy from early 20th century, who could add, subtract, work with fractions and even read, using his forelimbs to answer questions. The then recent works of Charles Darwin had further fuelled public interest in his abilities.
Hence, ______ ____ is a machine learning library which deals with adversarial attacks, a technique commonly used to unmask charlatan ML models, and demonstrate their brittleness in decision making.


One thought on “Summer QoTD #12

  1. Turned out to be a tough one.
    The answer is Clever Hans, also the name of a horse from Germany who could “solve” math problems and puzzles. It was later revealed that he did so by observing the reaction of the person asking the question, rather than through some deep understanding of the actual question.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s