Self-Explaining AI

Dan Elton joins us to discuss self-explaining AI. What could be better than an interpretable model? How about a model wich explains itself in a conversational way, engaging in a back and forth with the user.

We discuss the paper [Self-explaining AI as an alternative to interpretable AI](https://arxiv.org/abs/2002.05149) which presents a framework for self-explainging AI.

Guest

Dan Elton: Daniel Elton is a Staff Scientist at the National Institutes of Health working on applications of artificial intelligence to medical imaging. Much of his recent work is on improving automated measurements in CT and MRI scans using deep learning. He also works on theoretical issues surrounding robustness, explainability, and transparency. He is also the 2020 Foresight Fellow in AI at the Foresight Institute. He received a bachelor’s degree physics from Rensselaer Polytechnic Institute in 2010 and earned a Ph.D. in physics from Stony Brook University in 2016. During postdoctoral work at the University of Maryland he worked on deep learning for molecular property prediction,using deep generative models for molecular design, and applications of natural language processing to the chemical domain. His website is www.moreisdifferent.com/"