Explaining Explanations
Speaker: Leilani H. Gilpin, Massachusetts Institute of Technology
Abstract
There has recently been a surge of work in explanatory
artificial intelligence (XAI). This research area tackles the
important problem that complex machines and algorithms often cannot
provide insights into their behavior and thought processes. XAI
allows users and parts of the internal system to be more transparent,
providing explanations of their decisions in some level of
detail. These explanations are important to ensure algorithmic
fairness, identify potential bias/problems in the training data, and
to ensure that the algorithms perform as expected. However,
explanations produced by these systems are neither standardized nor
systematically assessed. In an effort to create best practices and
identify open challenges, in this talk, I describe and define the
foundational concepts of explainability and show how they can be used
to classify existing literature. I discuss why current approaches to
explanatory methods, especially for deep neural networks, are
insufficient. A review paper on this subject is available on arXiv
and as a conference proceeding [1].
[1] L. H. Gilpin, D. Bau, B. Z. Yuan, A. Bajwa, M. Specter and
L. Kagal, "Explaining Explanations: An Overview of Interpretability of
Machine Learning," 2018 IEEE 5th International Conference on Data
Science and Advanced Analytics (DSAA), Turin, Italy, 2018, pp. 80-89.
The slides can be downloaded here.
Bio
Leilani H. Gilpin is a PhD candidate in Electrical Engineering
and Computer Science at MIT, supervised by Prof. Gerald Jay Sussman
and funded by the Toyota Research Institute. She works on enabling
autonomous vehicles, and other autonomous machines, to explain
themselves. More information on her research interests can be seen at
[people.csail.mit.edu/lgilpin]. Before MIT, Leilani worked as a
research engineer at Palo Alto Research Center (PARC) focusing on
anomaly detection in healthcare. Leilani earned a M.S. in
Computational Mathematical and Engineering from Stanford University in
2013, and a B.S. in Mathematics (with honors), B.S. in Computer
Science (with highest honors), and a music minor from UC San Diego
in 2011.