This event has passed.

Exploiting metric structure for more accurate classification (Prof. Mike Izbicki)

Name: Exploiting metric structure for more accurate classification (Prof. Mike Izbicki)
Start: 2022-02-02T16:15:00-08:00
End: 2022-02-02T17:30:00-08:00
Location: Zoom meeting

February 2, 2022 @ 4:15 pm - 5:30 pm

Title: Exploiting metric structure for more accurate classification

Speaker: Mike Izbicki, Department of Mathematical Sciences, Claremont McKenna College

Abstract: Classification problems often have many semantically similar classes. For example, the famous ImageNet dataset contains classes for 80 different dog breeds, 40 different bird species, and 25 types of vehicles. This semantic structure can be formalized using a metric space, with semantic similarity of classes encoded by the distance function. In this talk, I’ll describe the “tree loss”, which is the first technique with provable performance guarantees for exploiting this metric structure. I’ll also show that the tree loss has better empirical performance than competing algorithms on image, text, and vector data.

Mike studies machine learning theory, focusing on applications to natural language and social media. He has been at CMC for 3 years now, where he teaches computer and data science classes. Prior to his academic career, Mike spent 7 years in the US Navy. Highlights include converting >10g of Uranium into pure energy as a nuclear submarine officer, and doing [redacted] for the NSA. After leaving the navy, Mike went to North Korea to teach computer science as part of an academic exchange program designed to improve relations between the US and North Korea. He earned his phd from UC Riverside.

Details

Date:: February 2, 2022
Time:: 4:15 pm - 5:30 pm
Event Category:: Colloquium

Organizer

: Christina Edholm

Venue

: Zoom meeting
: United States + Google Map