Lex Fridman Podcast

Pieter Abbeel: Deep Reinforcement Learning

Dec 16, 2018

42:56

Creator website

Highlights

AI Chapters

Episode notes

Is the Brain Really Modular?

01:42

The Future of AI Safety?

03:11

Are You Interested in Self Play?

03:37

Introduction

3min

Swinging a Racket, Is It Learnable?

2min

The Psychology of Interacting With Robots

2min

Is There a Signal That Could Be Used in Reinforcement Learning?

2min

RLS Reinforcement Learning

2min

Is RL Inefficient?

3min

The Decision of Doing a PhD

2min

Is There a Way to Bring Hierarchical Reasoning Together?

4min

Transfer Learning

3min

10.

Learning to Master Versus Learning to Generalize

2min

11.

Is the Brain Really Modular?

2min

12.

Are You a Mathematician or an Empirical Trial in an Error?

2min

13.

Are You Interested in Self Play?

4min

14.

Is Third Person Watching the Autonomous Driving Enough?

2min

15.

The Power of Simulation

2min

16.

The Future of AI Safety?

3min

17.

Is It Really Easy to Find Policies That Are Full of Kindness?

2min

18.

Do You Think It's Possible to Teach RL-Based Robot to Love a Human Being?

3min

Pieter Abbeel is a professor at UC Berkeley, director of the Berkeley Robot Learning Lab, and is one of the top researchers in the world working on how to make robots understand and interact with the world around them, especially through imitation and deep reinforcement learning. Video version is available on YouTube. If you would like to get more information about this podcast go to https://lexfridman.com/ai or connect with @lexfridman on Twitter, LinkedIn, Facebook, or YouTube where you can watch the video versions of these conversations.