Beating Stochastic and Adversarial Semi-bandits Optimally and Simultaneously

01/25/2019
by   Julian Zimmert, et al.
0

We develop the first general semi-bandit algorithm that simultaneously achieves O( T) regret for stochastic environments and O(√(T)) regret for adversarial environments without knowledge of the regime or the number of rounds T. The leading problem-dependent constants of our bounds are not only optimal in some worst-case sense studied previously, but also optimal for two concrete instances of semi-bandit problems. Our algorithm and analysis extend the recent work of (Zimmert & Seldin, 2019) for the special case of multi-armed bandit, but importantly requires a novel hybrid regularizer designed specifically for semi-bandit. Experimental results on synthetic data show that our algorithm indeed performs well uniformly over different environments. We finally provide a preliminary extension of our results to the full bandit feedback.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset

Sign in with Google

×

Use your Google Account to sign in to DeepAI

×

Consider DeepAI Pro