The Robot Brains Podcast

Chelsea Finn: how to build AI that can keep up with an always changing world

Mar 22, 2023

01:02:05

Creator website

Highlights

AI Chapters

Episode notes

Is There a Difference Between Machine Learning and Robotics?

03:38

How Can Crowdsourcing Improve Reward Learning?

02:22

Is There a Way to Automate the Editing Process?

01:38

Introduction

3min

What Can AI Not Do Today?

2min

Is There a Distribution Shift?

3min

How to Make a Targeted Edit to a Model

3min

Is There a Way to Automate the Editing Process?

2min

Is There a Difference Between Machine Learning and Robotics?

4min

Robotics and Machine Learning - Is There a Challenge?

2min

Learning Robotics From Offline Data

2min

Can Machine Learning Generalize Beyond the Training Distribution?

3min

10.

Single Life Reinforcement Learning - I Hadn't Seen It Before

2min

11.

Reward Functions in Deep Reactor Learning

3min

12.

How Can Crowdsourcing Improve Reward Learning?

2min

13.

Is Covering the Whole Space Really Intractable?

2min

14.

Are We Just Not Trading Long Enough?

3min

15.

How Do You Collect Large Data Sets?

3min

16.

How Did the Mammal Paper Come About?

3min

17.

Meta Learning

3min

18.

Is There a Recursive Form of Meta Learning?

2min

19.

Robotics, What's Your Vision for the Next Five to Ten Years?

2min

20.

How Did You Become Into Artificial Intelligence as a Kid?

2min

21.

What Triggered Your Focus on AI?

3min

22.

Are There Any Productivity Tips You Can Share?

3min

23.

How to Recruit a PhD Student?

2min

24.

Is There a Future in Robotics?

2min

Chelsea Finn joins Host Pieter Abbeel to discuss distribution shift, meta-learning, editing LLMs, single-life RL, and what can AI not (yet) do today.

Subscribe to the Robot Brains Podcast today | Visit therobotbrains.ai and follow us on YouTube at TheRobotBrainsPodcast and Twitter @therobotbrains.

Hosted on Acast. See acast.com/privacy for more information.

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.