Tue 2/25, 4:30pm, DSI Talk: Cameron Musco (UMass) - Leveraging Randomized Algorithms in Modern ML
Sarah Wu
Hi All,
Please join us, on Tuesday 2/25 at 4:30pm in SCCE E110, for a talk part of the Data Science Initiative speaker series.
Speaker: Cameron Musco (UMass)
Title: Leveraging Randomized Algorithms in Modern Data Science and Machine Learning
Abstract: Randomized algorithms are computer programs that make random decisions as part of their logic. In this talk, I will highlight the central role that these algorithms play in large-scale data processing and machine learning. I will start by discussing how very simple and intuitive random sampling methods have been critical in driving the rise of modern machine learning, including deep neural networks and generative AI. I will then discuss techniques that attempt to go beyond what is achievable with vanilla random sampling. I will introduce the ideas of importance sampling and random linear sketching, as ways of coping with highly skewed data distributions or potentially worst-case input instances.
Light fancy refreshments will be served before the talk.
Link to calendar file: https://www.amherst.edu/mm_calendar_ical/925581/event.ics (Note: The event is at 4:30 PM!)
Flyer attached.
Feel free to spread the word to faculty, staff, and students.
All our events are listed on https://www.amherst.edu/about/science_at_amherst/data-science-initiative
We hope to see you there!
