Beating Bandits in Gradually Evolving Worlds

  title={Beating Bandits in Gradually Evolving Worlds},
  author={Chao-Kai Chiang and Chia-Jung Lee and Chi-Jen Lu},
Consider the online convex optimization problem, in which a player has to choose actions iteratively and suffers corresponding losses according to some convex loss functions, and the goal is to minimize the regret. In the full-information setting, the player after choosing her action can observe the whole loss function in that round, while in the bandit setting, the only information the player can observe is the loss value of that action. Designing such bandit algorithms appears challenging, as… CONTINUE READING

From This Paper

Topics from this paper.
7 Citations
15 References
Similar Papers


Publications referenced by this paper.
Showing 1-10 of 15 references

Optimal algorithms for online convex optimization with multi-point bandit feedback

  • Alekh Agarwal, Ofer Dekel, Lin Xiao
  • In COLT,
  • 2010
Highly Influential
15 Excerpts

Similar Papers

Loading similar papers…