CBMM Panel Discussion: Are Deep Nets "Just" Associative Memories?

Speaker: Multiple speakers

Date: Tuesday, March 09, 2021

Time: 4:00 PM to 5:30 PM Note: all times are in the Eastern Time Zone

Public: Yes

Location: Virtual | Webinar

Event Type: Seminar

Room Description: This panel discussion will be hosted remotely via Zoom.

Host: Prof. Tomaso A. Poggio

Contact: Kathleen Sullivan, kdsulliv@csail.mit.edu

Relevant URL: https://cbmm.mit.edu/news-events/events/cbmm-panel-discussion-are-deep-nets-just-associative-memories

Speaker URL: None

Speaker Photo:

Reminders to: seminars@csail.mit.edu

Reminder Subject: TALK: CBMM Panel Discussion: Are Deep Nets "Just" Associative Memories?

Panelists: Profs. Christos Papadimitriou (Columbia), Tomaso A. Poggio (CBMM, MIT) and Santosh Vempala (Georgia Tech)
Moderator: TBA

Abstract: About fifty years ago, holography was proposed as a model of associative memory. Associative memories with similar properties were soon after implemented as simple networks of threshold neurons by Willshaw and Longuet-Higgins. It turns out that the recurrent Willshaw networks were very similar to today's deep nets. Thinking about deep learning in terms of associative networks memories a more realistic and sober perspective on the promises of deep learning and on its role in eventually understanding human intelligence.

This panel discussion will be hosted remotely via Zoom.
Zoom link: https://mit.zoom.us/j/99556310473

Research Areas:
Algorithms & Theory, AI & Machine Learning

Impact Areas:

See other events that are part of the Brains, Minds and Machines Seminar Series 2020 - 2021.

Created by Kathleen Sullivan at Tuesday, March 02, 2021 at 3:23 PM.