Hongsuk Benjamin Choi
@redstone_hong
We saw exactly this in the ABC paper (abc.bot) on real robots. Across 16 checkpoints, validation flow loss was uncorrelated with real-world success, while validation action error had the strongest correlation. Fig 8, if you're curious.
abc.botScalable Behavior Cloning with Open Data, Training, and Evaluation
Seohong Park@seohong_park · Aug 24Behavioral cloning mystery
seohong.me/blog/behaviora…
I wrote a new blog post about "mysteries" in behavioral cloning that appear with real-world robot data (e.g., overfitting is "good"). I also tried to demystify them and shared my thoughts!
seohong.me/blog/behaviora…
I wrote a new blog post about "mysteries" in behavioral cloning that appear with real-world robot data (e.g., overfitting is "good"). I also tried to demystify them and shared my thoughts!
Open quoted post →
2 96