The main problem of deep learning is that most applications of this technique are black boxes and hence is difficult to explain why a decision was taken. This is also true in most ML techniques. However, in many applications that affect people, the decision must be explained for legal (e.g., GDPR, the new privacy law of the European Union) and/or ethical reasons (e.g., gender or racial bias). We cover the main challenges and we give examples of current techniques to explain ML models, ending with best practices and an analysis of the possible future.
After this presentation, a new flexible data science learning program that can end in a Masters in Data Science or Computer Science will be presented and, of course, explained.
Our Speaker: Dr. Ricardo Baeza-Yates
View his profile here: http://www.northeastern.edu/siliconvalley/faculty/