A Conversation Between Terry Sejnowski & Stephen Wolfram (February 14, 2023) [Part 2]

1

Introduction

00:00 • 2min

2

Is It Possible to Reduce a Loss in a Neural Network?

02:16 • 2min

3

The Origin of the Back Prop Idea

03:47 • 3min

4

Using Backpropagation in a Network

06:23 • 2min

5

How to Make a Learning Algorithm Work for You

08:04 • 4min

6

The Biola Jones Face Detector

11:44 • 2min

7

How to Get 10% Better Performance Out of Back Propagation

13:43 • 5min

8

The Brain Will Be a Source of Inspiration for Decades to Come

18:29 • 2min

9

Neural Nets - A History of Computer Science

20:11 • 2min

10

Is There a Future for DARPA?

22:05 • 2min

11

The Image Net Competitions, Right?

24:06 • 2min

12

Is That an Apocryphal Story?

25:44 • 2min

13

GPUs, When Did They Come on the Scene?

27:56 • 2min

14

Is That the on Lecun's Limit?

29:43 • 2min

15

Caltech - The CNS Program

31:36 • 2min

16

What if You Can Change the Hamiltonian?

33:50 • 2min

17

Is That the Original Hopfield Network Idea?

35:29 • 2min

18

Is It More, It Is Better?

37:12 • 3min

19

Is There a History of Machine Learning?

40:31 • 3min

20

Is It a Bug or a Feature?

43:49 • 2min

21

The World Works on a Millisecond Time Scale

45:28 • 3min

22

What's the Next Stage?

48:27 • 2min

23

The Reverse Turing Test

50:17 • 2min

24

Is It a Meta-Algorithm?

52:06 • 3min

25

How Much Data Do You Need?

55:26 • 2min

26

Is It Possible to Represent That With A?

57:29 • 2min

27

Do You Know How to Measure a Brain's Behavior?

59:45 • 2min

28

How Many Lizard Goddess Spots Are There?

01:01:23 • 3min

29

The Dimensional Law of Motion

01:04:10 • 3min

30

Then, You Know, You've Got a Problem, Right?

01:07:31 • 5min

31

How Difficult Is This?

01:12:30 • 2min

32

What Does Solve the Problem Mean?

01:14:02 • 3min

33

Learning a Black Box Function

01:16:50 • 3min

34

How Randomizing Glasses Can Improve Learning Algorithms

01:19:57 • 2min

35

What Kind of Neural Networks Did Robert Heck-Nealsons Use?

01:21:52 • 2min

36

The Human Brain Project

01:24:10 • 2min

37

Are There Algorithmic Developments of Back Propagation?

01:26:17 • 3min

38

Is There a Narrative Scientific Story of How Neural Nets Work?

01:29:35 • 3min

39

The Future Is Limitless, Right?

01:32:39 • 2min

40

Neural Nets - The Only Existence Proof That Any Problem Could Be Solved, Right?

01:34:33 • 2min

41

Chat GPT

01:36:05 • 2min

42

What's the Outer Loop of Human Language?

01:37:50 • 2min

43

The Architecture of the Brain Supports Execution of Computational Language

01:40:15 • 2min

44

The History of Quantum Mechanics

01:42:01 • 2min

45

Neural Networks - The Tree Grows From One Trunk

01:44:14 • 3min