The Inside View

Jesse Hoogland on Developmental Interpretability and Singular Learning Theory

Jul 6, 2023
Ask episode
Chapters
Transcript
Episode notes
1
Introduction
00:00 • 2min
2
The Importance of Being Ethical
01:36 • 2min
3
AI Safety: A Turning Point
03:07 • 2min
4
The Singular Learning Theory
04:43 • 2min
5
The Importance of Singular Learning Theory in AI Safety
06:38 • 2min
6
The Importance of Singular Learning Theory
08:28 • 2min
7
Singular Learning Theory: A New Approach to Machine Learning
10:58 • 5min
8
The Importance of Face Transitions in AI
15:29 • 2min
9
The Effects of Face Transitions on Neural Networks
17:13 • 2min
10
The Importance of Interpretability
18:52 • 2min
11
The Universality Hypothesis for Learning Online
20:48 • 2min
12
The Problem With Interpretability in Large Systems
22:56 • 3min
13
The Evolution of the Human Brain
25:29 • 2min
14
The Prosaic Boring Answer
27:39 • 2min
15
The Problem With the Manhattan Project
29:58 • 2min
16
The Future of Artificial Intelligence
31:45 • 2min
17
The Importance of Alignment in AI
33:41 • 2min
18
The Myth of Creativity
36:08 • 2min
19
Singular Learning Theory: A Pathway to AI Safety
38:01 • 2min
20
Singular Learning Theory Developmental Interpretability
39:39 • 4min