Deanna Needell, UCLA

Dunn Family Endowed Chair in Data Theory
Executive Director, Institute for Digital Research and Education (IDRE)

Abstract: In this talk I will start by sharing a bit about my own story, and some conjectures about why numbers for women in STEM have been historically low. Then, we will delve into the following questions, all of which center around the concept of large-scale quantities, objects, and data. First, we ask "Does Alexa really know how likely it is that two people shuffle a deck of cards in the same order? -- wait, do I even know this?". To grapple with this question we will embark on a thought experiment that takes us around the world. Next we ask "Why is it hot in dimensional space and what does that even mean?", which will include some surprising geometric results. Lastly, we consider "Is there such thing as a 'clear-box' machine learning method, and why is this important?". This will include recent work in linear algebraic approaches to machine learning and new applications to which we have applied these methods. No prior expertise is assumed, and all are welcome!