An Active Approach for Model Interpretation

Jialin Lu, Martin Ester

NeurIPS 2019 workshop on Human-centric machine learning (HCML2019)

An Active Approach for Model Interpretation

Every computer science project has a two-fold reason for why we do it. The first is why it matters to the real world, in terms of impact and consequences, and the second reason is its interestingness from a computational viewpoint (after all, we do computer science)

This project is being actively development and have a milestone (A NeurIPS workshop paper) and I am still on further inverstigationss. Feel free to contact me if you're interesting.

• 1. Why it matters to the real world? Model interpretation for pattern discovery.
• 2. Why it is interesting from a computational viewpoint.

Other Preliminary draft see here:

• Version on August 27th, 2019
• Version on June the 4th, 2019