Improved Regret Bounds for Oracle-Based Adversarial Contextual Bandits

Authors: Vasilis Syrgkanis, Haipeng Luo, Akshay Krishnamurthy, Robert E. Schapire

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We propose a new oracle-based algorithm, BISTRO+, for the adversarial contextual bandit problem...Our algorithm is computationally efficient, assuming access to an offline optimization oracle, and enjoys a regret of order O((KT) 2 3 (log N) 1 3 )...Our algorithm and regret bound are based on a novel and improved analysis of the minimax problem that arises in the relaxation-based framework of Rakhlin and Sridharan [7].
Researcher Affiliation Collaboration Vasilis Syrgkanis Microsoft Research vasy@microsoft.com Haipeng Luo Microsoft Research haipeng@microsoft.com Akshay Krishnamurthy University of Massachusetts, Amherst akshay@cs.umass.edu Robert E. Schapire Microsoft Research schapire@microsoft.com
Pseudocode Yes Algorithm 1 BISTRO+ Algorithm 2 Computing q t (ρt)
Open Source Code No The paper does not provide any statement or link regarding the availability of open-source code for the described methodology.
Open Datasets No The paper is theoretical and focuses on algorithm design and regret bounds. It does not use specific publicly available datasets for empirical training or evaluation. It mentions 'contexts are drawn i.i.d. from a distribution D', implying a theoretical model rather than a specific dataset.
Dataset Splits No The paper is theoretical and does not describe empirical experiments with dataset splits for training, validation, or testing.
Hardware Specification No The paper is theoretical and does not mention any specific hardware used for experiments.
Software Dependencies No The paper describes an algorithm and its theoretical bounds but does not list any specific software dependencies with version numbers.
Experiment Setup No The paper is theoretical and focuses on algorithm design and analysis. It does not provide details about an experimental setup, such as hyperparameter values or training configurations.