Improved Regret Guarantees for Online Smooth Convex Optimization with Bandit Feedback.
Ankan SahaAmbuj TewariPublished in: AISTATS (2011)
Keyphrases
- convex optimization
- online convex optimization
- online learning
- interior point methods
- low rank
- total variation
- primal dual
- convex optimization problems
- bandit problems
- online algorithms
- regret bounds
- lower bound
- loss function
- basis pursuit
- convex relaxation
- norm minimization
- higher order
- convex formulation
- image restoration
- semidefinite program
- operator splitting
- upper confidence bound
- denoising