A measurements-based approach to machine learning

Lecture

Speaker(s)

Gerald Friedland

Adjunct Assistant Professor, EECS, UC Berkeley, BIDS Faculty Affiliate

BIDS Faculty Affiliate Gerald Friedland is an Adjunct Assistant Professor of Electrical Engineering and Computer Sciences at UC Berkeley, and the co-founder and CTO of Brainome. His work focusses on large-scale machine learning for multimedia retrieval, and he has also worked in privacy and privacy education. He has published more than 200 peer-reviewed articles in conferences, journals, and books and also co-authored the textbook Multimedia Computing. Dr. Friedland received his doctorate in computer science from Freie Universitaet Berlin, Germany.

Wednesday, September 1, 2021

CITRIS Research Exchange presents
A Measurements-Based Approach to Machine Learning
Date: September 1, 2021
Time: 12:00–1:00 PM Pacific
Speaker: Gerald Friedland, Adjunct Assistant Professor, Electrical Engineering and Computer Science, UC Berkeley

Abstract: Every field of science and engineering starts with measurements. When working on machine learning problems, modern data science often relies more on computation (let’s throw more GPUs at the problem) and guesswork (let’s see if we can modify Alexnet for our specific problem) than on any kind of measurements. In this talk, I will discuss an approach to supervised machine learning that is rooted in information-theoretic measurements. I will explain how this approach is especially interesting for high dimensional problems that, without measurements, suffer from “the curse of dimensionality”. I will explain the fundamentals of how measurements-based machine learning works, and also explore how the approach can be applied to solve real-world problems in CITRIS and beyond.

The CITRIS Research Exchange series delivers fresh perspectives on information technology and society from distinguished academic, industry, and civic leaders. Free and open to the public, this series highlights leading voices on societal-scale research issues. Each seminar takes place on Wednesdays from 12:00 pm to 1:00 pm PT. Have a suggestion for a great speaker? Please use this form to suggest potential speakers for our series.

Share this video