Brain Inspired

BI 129 Patryk Laurent: Learning from the Real World

Mar 2, 2022

01:21:01

Creator website

Highlights

AI Chapters

Episode notes

The Impetus Behind the Architecture, Right?

01:37

Ai Robotics - Can They Generalize Well?

02:32

Your View of Intelligence

01:44

Introduction

3min

Posttock - Graduate School and Your Academic Experience?

2min

The Challenges of Getting a PhD in Computer Science

2min

What Was Your First Job?

2min

Do You Have a Diverse Background?

2min

Do You Think It's Healed You at All?

4min

Is There Room for the Scribbled Shape People and Scibbles?

2min

How Do People Measure Intelligence?

2min

Your View of Intelligence

2min

10.

Building Robotic Brains

3min

11.

How to Predict the Next Time Step in a Robotic Environment?

3min

12.

The Architecture That Ugys Uses Is Similar to Multiple Other Architectures

2min

13.

The Impetus Behind the Architecture, Right?

2min

14.

Is There Too Much Recurrence?

2min

15.

Using Python to Train a Deep Learning Architecture, It Just Keeps Getting Better

2min

16.

Transformers - What Are Your Thoughts?

3min

17.

Are Semantics Still Missing?

3min

18.

Are Transformers Teaching Us Anything About Our Intelligence?

2min

19.

Machine Learning Can Help Us Understand Our Intelligence, Right?

2min

20.

Is Human Level Performance the Goal?

2min

21.

Does Human Level Performance Matter?

1min

22.

Is It a Failure of the A?

2min

23.

Ai, I Don't Care About You, Patrick, Right?

2min

24.

Ai Robotics - Can They Generalize Well?

3min

25.

How Can the Brain Better Inform Ai?

4min

26.

Is There a Sleep Related Algorithm?

3min

27.

Can We Safely Ignore Astrocites?

3min

28.

Is There a Need for a Little Read Right Memory in Artificial Networks?

3min

29.

The Impossible Questions

2min

30.

Is It Easier to Start Your Own Lab?

3min

31.

Teaching

2min

32.

Do You Feel Like You Have Stepped Out of Your Comfort Zone?

2min

33.

Is Coating Slash Soft Ware Development Required in Neuro Science?

4min

34.

I'm a Chess Player, Right?

3min

Support the show to get full episodes, full archive, and join the Discord community.

Patryk and I discuss his wide-ranging background working in both the neuroscience and AI worlds, and his resultant perspective on what's needed to move forward in AI, including some principles of brain processes that are more and less important. We also discuss his own work using some of those principles to help deep learning generalize to better capture how humans behave in and perceive the world.

Patryk's homepage.
Twitter: @paklnet.
Related papers
- Unsupervised Learning from Continuous Video in a Scalable Predictive Recurrent Network.

0:00 - Intro 2:22 - Patryk's background 8:37 - Importance of diverse skills 16:14 - What is intelligence? 20:34 - Important brain principles 22:36 - Learning from the real world 35:09 - Language models 42:51 - AI contribution to neuroscience 48:22 - Criteria for "real" AI 53:11 - Neuroscience for AI 1:01:20 - What can we ignore about brains? 1:11:45 - Advice to past self