Oracle-Efficient and Parameter-Free Agnostic Smoothed Online Learning
A parameter-free Follow-The-Perturbed-Leader method gets near-optimal agnostic smoothed online regret without knowing the base measure.
1 stories, newest first · one card per story
A parameter-free Follow-The-Perturbed-Leader method gets near-optimal agnostic smoothed online regret without knowing the base measure.
Today 459 readers · 514 views · 154 reads · all time 187K views