News & Event

Seminar

News & Event

Seminar

Generative Modeling with Succinct Representation via Wyner’s Common Information

Subject

Generative Modeling with Succinct Representation via Wyner's Common Information

Date

Jan. 18 (Wednesday) 11 AM

Speaker

Dr. Jongha Ryu, postdoc associate at MIT

Place

N1, Room 201

Overview:

O Speaker:  Dr. Jongha Ryu, postdoc associate at MIT

O Title: Generative Modeling with Succinct Representation via Wyner’s Common Information

O Date: Jan. 18 (Wednesday)

O Start Time: 11 AM

O Venue: N1, Room 201

 

O Abstract:

In this talk, I will introduce an information-theoretic approach to develop a deep generative model for two correlated high-dimensional random vectors such as images and captions, for solving machine learning tasks such as joint and conditional sample generation and cross-domain retrieval. Based on Wyner’s common information from network information theory, I will motivate a new bimodal generative model framework that seeks a succinct common representation, and describe how to train the proposed model systematically. 

 
O Bio:
Jongha (Jon) Ryu is currently a postdoctoral associate at Research Laboratory of Electronics, MIT, Cambridge, MA, United States. He received the Ph.D. and M.S. degrees in electrical and computer engineering from UC San Diego, La Jolla, CA, United States, in 2022 and 2018, respectively, and received the B.S. degrees in electrical and computer engineering and mathematical science from Seoul National University, Seoul, South Korea, in 2015. He was a recipient of the Kwanjeong Scholarship for graduate study from 2015 to 2020. His research interest broadly lies at the intersection of information theory and machine learning.

 

Profile: