Manipulating and Measuring Model Interpretability — mentioned by lalaland1125

Thu, 11 Jul 2019 03:23:24 GMT

"Interpretability" is probably one of the most misunderstood topics in ML. The key with interpretability is that it's actually an Human Computer Interaction (HCI) problem, not a statistics or math problem. The problem is that people have particular tasks that they need to accomplish and the goal of interpretability techniques is to help them solve those tasks. This is somewhat similar thematically to how a keyboard layout can help someone achieve their typing speed goals. The two main tasks that interpretability is usually geared towards is model error detection and hypothesis generation. In…

Paper and discussion history. A mention is not an endorsement.

Manipulating and Measuring Model Interpretability · HackerPapers discussions

Manipulating and Measuring Model Interpretability — mentioned by lalaland1125