Phillip Isola - Embodied Intelligence Seminar

Speaker: Phillip Isola , MIT

Date: Wednesday, February 19, 2020

Time: 2:00 PM to 3:30 PM Note: all times are in the Eastern Time Zone

Public: Yes

Location: Seminar Room D463 (Star)

Event Type:

Room Description:

Host: Ferran Alet, MIT

Contact: Ferran Alet, alet@csail.mit.edu

Relevant URL:

Speaker URL: None

Speaker Photo:
None

Reminders to: csail-related@mit.edu

Reminder Subject: TALK: Phillip Isola - Embodied Intelligence Seminar

Title:
Emergent Intelligence: getting more out of agents than you bake in

Abstract:
One of the most satisfying things in AI research is to get more out than you put in. I will talk about three projects where we got something surprising out, beyond what was explicitly baked in. First, a simple inductive bias for unsupervised learning discovers “semantics”, despite that no labels are used during training. Second, physically plausible “videos” emerge in the latent space of a generative model, despite that the model was only fit to static images. Third capable and robust “creatures” self-assemble from a collection of primitive limbs, despite having no centralized controller. I will suggest that a goal in artificial intelligence is to get the most out while baking the least in.

Bio:
Phillip Isola is the Bonnie & Marty Tenenbaum Career Development Chair Assistant Professor in EECS at MIT. He completed his Ph.D. in Brain & Cognitive Sciences at MIT, advised by Edward Adelson, then spent two years as a postdoc at UC Berkeley, under Alexei Efros, and one year as a visiting research scientist at OpenAI. His work focuses on why we represent the world the way we do, and how we can replicate these abilities in machines.

After the talk we will have a social event with food.

Research Areas:
AI & Machine Learning, Graphics & Vision, Robotics

Impact Areas:

This event is not part of a series.

Created by Ferran Alet Email at Tuesday, February 18, 2020 at 10:25 PM.