This quarter we'll talk about AI safety, explainable AI, human-in-the-loop AI, and ethical challenges for AI. We’ll have some guest speakers, but much of the seminar will be spent discussing some recent papers on the topic.
uw-aimailing list. Subscribe here.
|10/5||Reading||Antoine Bosselut||One Hundred Year Study on Artificial Intelligence: Report of the 2015-2016 Study Panel|
|10/12||Reading||Ariel Holtzman||Concrete Problems in AI Safety|
|10/19||Cancelled||because of overlap with the Affiliates Day.|
|10/26||Reading||Chris & Nicholas||
Stealing Machine Learning Models
via Prediction APIs
|11/02||Reading||Gagan||“Why Should I Trust You?” Explaining the Predictions of Any Classifier|
|11/09||Talk||Oren Etzioni (AI2)||
Designing AI systems that obey our laws and values
Relevant reading on Semantic Scholar
|11/16||Reading||Colin Lockard||The forces that drove this election’s media failure are likely to get worse|
|11/30||Talk||Ryan Calo (UW School of Law)||
There is a blind spot in AI research
Relevant Nature article