Generally Intelligent Jacob Steinhardt, UC Berkeley: Machine learning safety, alignment and measurement
Jun 18, 2021
Chapters
Transcript
Episode notes
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Introduction
00:00 • 6min
How Do You End Up Working on the Future of Machine Learning?
05:37 • 5min
The Work on Competationally Based Reasoning Was Hopeful
10:41 • 4min
How Do You Scale With the Dimension of the Problem?
14:15 • 2min
How to Learn Non-Adversarial Data Sets
16:18 • 3min
The Heroes of the Day
19:20 • 2min
Measurement for Results
21:05 • 5min
Is There a Measure for Robustness or Capabilities?
25:50 • 2min
Why Is It So Hard to Be Calebrated?
28:02 • 5min
The History of M L Models
33:02 • 3min
What Papers Have Impacted You Most?
35:43 • 4min
Is There a Filter in Your Information?
39:22 • 2min
Hedgehogs or Foxes?
41:07 • 4min
Hedgehogs Are the Victims of Their Own Success
44:49 • 2min
Do You Feel Like There Are Underrated Approaches or Techniques?
46:32 • 5min
Do You Really Need to Critique Other People's Ideas?
51:35 • 2min
How to Create a Great Environment for Collaboration in a Group Discussion
53:11 • 4min
What Makes E Great, Mister Te Grange?
57:22 • 3min