Linear Bandits with Feature Feedback

Authors: Urvashi Oswal, Aniruddha Bhargava, Robert Nowak5331-5338

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate the performance of the algorithm with synthetic and real human-labeled data.
Researcher Affiliation Collaboration 1Oracle Labs, Redwood Shores, CA 2University of Wisconsin-Madison, Madison, WI
Pseudocode Yes Algorithm 1 OFUL; Algorithm 2 Feature Feedback OFUL (FF-OFUL)
Open Source Code No The paper does not provide any explicit statements about open-sourcing the code for the described methodology or links to a code repository.
Open Datasets Yes We use the 20Newsgroup (20NG) dataset from (Lang 1995).
Dataset Splits No The paper describes the datasets used (synthetic and 20Newsgroup) but does not provide specific train/validation/test splits (e.g., percentages, counts, or references to predefined splits).
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU models, CPU types, memory) used for running the experiments.
Software Dependencies No The paper does not specify any software dependencies with version numbers.
Experiment Setup Yes For OFUL the ridge parameter (λ) is tuned from {2i}10 i= 7 to pick the one with best performance. All the tuned parameters selected for OFUL were strictly inside this range (for d = 40, k = 5 , λ = 2 5 and for d = 103 (Newsgroup), λ = 28). For FF-OFUL, the remarkable feature is that it does not require parameter tuning so λ = 1 for all experiments.