First Principles Manufacturing
The FPM Breakdown
End-to-End Robot Learning or Motion Planning? The Bet on Where Robotics Ends Up
0:00
-9:58

End-to-End Robot Learning or Motion Planning? The Bet on Where Robotics Ends Up

The FPM Breakdown, episode 11: amortize vs defer in plain English, why neither layer can be dropped, and why a success rate isn’t a certificate.

A breakdown of Amortize or Defer, from First Principles Manufacturing, a newsletter by Yuanwei Ma at the intersection of AI, robotics, manufacturing and US–China.

Should a robot learn everything end to end, or plan its moves on the spot? The piece answers with a bet, probabilities attached: a hybrid wins, and neither layer can be dropped. We translate amortize and defer into plain English, take apart why both are necessary, why a success rate is not a certificate, where the frontier really is, and a precedent from factory defect inspection that already played this out.

Read the full article: Amortize or Defer

Voices are AI-generated. The facts come from the original article by Yuanwei Ma; the interpretation is the hosts'.QQQ

Chapters

  • 0:00 End-to-end learning or planning? A bet you can grade

  • 1:10 Amortize vs defer, in plain English

  • 2:42 Why neither layer can be dropped, and what a certificate is

  • 4:34 Where the frontier is: search is mainstream, verification is not

  • 6:13 A precedent: how defect inspection settled in two years

  • 8:08 Isn't this just a safety layer? Testable predictions

Discussion about this episode

User's avatar

Ready for more?