Paper bounds private sparse reward estimation from preference feedback
Sample-DP and Label-DP rates depend on sparse features s* rather than full model size d.
Stories per day · 30 days
1 stories, newest first · one card per story
1 stories, newest first · one card per story
Sample-DP and Label-DP rates depend on sparse features s* rather than full model size d.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views