Adversarial Dueling Bandits

10/27/2020
by   Aadirupa Saha, et al.
0

We introduce the problem of regret minimization in Adversarial Dueling Bandits. As in classic Dueling Bandits, the learner has to repeatedly choose a pair of items and observe only a relative binary `win-loss' feedback for this pair, but here this feedback is generated from an arbitrary preference matrix, possibly chosen adversarially. Our main result is an algorithm whose T-round regret compared to the Borda-winner from a set of K items is Õ(K^1/3T^2/3), as well as a matching Ω(K^1/3T^2/3) lower bound. We also prove a similar high probability regret bound. We further consider a simpler fixed-gap adversarial setup, which bridges between two extreme preference feedback models for dueling bandits: stationary preferences and an arbitrary sequence of preferences. For the fixed-gap adversarial setup we give an Õ((K/Δ^2)logT) regret algorithm, where Δ is the gap in Borda scores between the best item and all other items, and show a lower bound of Ω(K/Δ^2) indicating that our dependence on the main problem parameters K and Δ is tight (up to logarithmic factors).

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset