School of Information Sciences Undergrad MASTER CALENDAR

Back to Listing

The iSchool offers a number of events related to career and professional development, technology and information talks, research seminars, field trips, alumni panels, socials, and more. We also promote relevant opportunities on and around campus. 

We encourage students to also visit additional calendars and websites:

HandShake EventsResearch ParkNCSATechnology Entrepreneur Center (TEC),

The Career Center, Office of Undergraduate Research, Leadership Center, Siebel Center for Design,

Office of Technology Management, Center for Innovation in Teaching & Learning,

Applied Technologies for Learning in the Arts & Sciences.

National & International Scholar's Programs, Student Wellness

iSchool Calendars: Study Abroad Hours, iSchool Events, Non-iSchool Events

BSIS ICT Sessions, Express Advising

Blue Waters Webinar: "The Geometry of Data: Tackling the Challenges of Machine Learning at Scale"

Event Type
Conference/Workshop
Sponsor
NCSA
Location
webinar
Virtual
wifi event
Date
Oct 6, 2020   10:00 - 11:30 am  
Speaker
Aaron Saxton, NCSA
Registration
Registration
Views
13
Originating Calendar
Blue Waters events

The Blue Waters project at the University of Illinois is offering a webinar on "The Geometry of Data: Tackling the Challenges of Machine Learning at Scale" by Aaron Saxton on Tuesday, October 6, 2020 from 10:00-11:30 a.m. (Central Time)/11:00 a.m. - 12:30 p.m. (Eastern Time).

Please register to participate.

Abstract: There are many ways to parallelize computational workflows on HPC systems like Blue Waters and they all come with risks and benefits. The holy grail of paralyzing ML workflows is distributed training. In this process the model is copied, data is partitioned, and both are loaded onto multiple nodes to increase the scale at which we can train. In this talk, we start with the hypothesis that all data can be embedded on some lower dimension manifold. Indeed, this is called the geometric interpretation of data. Since gradient methods are the primary algorithms to optimize ML models on training data, by using the geometric interpretation we are able to visualize and gain insight to the challenges optimization faces. In particular we will explore how this expresses while training at scale. There is no good general theory of ML training, but this will give practitioners some intuitive tools to improve their models to push the limits of scaling.

link for robots only