undefined | Better HN

0 pointsmirker3y ago0 comments

Many of these scaling patterns are logarithmic with respect to data size. You can only double the dataset size so many times that it’s really not clear the scaling will continue.

0 comments

lostmsu3y ago

Low data modes are also progressing quite fast. There is Dreamer and more recent papers based on RL in learned world models.

j / k navigate · click thread line to collapse

0 pointsmirker3y ago0 comments

Many of these scaling patterns are logarithmic with respect to data size. You can only double the dataset size so many times that it’s really not clear the scaling will continue.

0 comments

lostmsu3y ago

Low data modes are also progressing quite fast. There is Dreamer and more recent papers based on RL in learned world models.

j / k navigate · click thread line to collapse