Corpus ID: 173990358

Model selection for contextual bandits

@inproceedings{Foster2019ModelSF,
  title={Model selection for contextual bandits},
  author={Dylan J. Foster and Akshay Krishnamurthy and Haipeng Luo},
  booktitle={NeurIPS},
  year={2019}
}
  • Dylan J. Foster, Akshay Krishnamurthy, Haipeng Luo
  • Published in NeurIPS 2019
  • Computer Science, Mathematics
  • We introduce the problem of model selection for contextual bandits, wherein a learner must adapt to the complexity of the optimal policy while balancing exploration and exploitation. Our main result is a new model selection guarantee for linear contextual bandits. We work in the stochastic realizable setting with a sequence of nested linear policy classes of dimension $d_1 < d_2 < \ldots$, where the $m^\star$-th class contains the optimal policy, and we design an algorithm that achieves $\tilde… CONTINUE READING

    Topics from this paper.

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 12 CITATIONS

    Model Selection in Contextual Stochastic Bandit Problems

    VIEW 6 EXCERPTS
    CITES BACKGROUND
    HIGHLY INFLUENCED

    Hedging the Drift: Learning to Optimize under Non-Stationarity

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Comparator-adaptive Convex Bandits

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Corralling Stochastic Bandit Algorithms

    VIEW 2 EXCERPTS
    CITES RESULTS & BACKGROUND

    Open Problem: Model Selection for Contextual Bandits

    VIEW 2 EXCERPTS
    CITES METHODS & BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 61 REFERENCES

    Learning to Optimize under Non-Stationarity

    VIEW 2 EXCERPTS

    Contextual Bandit Learning with Predictable Rewards

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL