A Practical Semi-Parametric Contextual Bandit
Authors: Yi Peng, Miao Xie, Jiahao Liu, Xuying Meng, Nan Li, Cheng Yang, Tao Yao, Rong Jin
IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on synthetic data as well as a real dataset from one of the largest e-commercial platforms demonstrate the superior performance of our algorithm. |
| Researcher Affiliation | Collaboration | Yi Peng1 , Miao Xie1 , Jiahao Liu1 , Xuying Meng2 , Nan Li1 , Cheng Yang1 , Tao Yao1 and Rong Jin1 1Alibaba Group, Hang Zhou, China 2Institute of Computing Technology, Chinese Academy of Sciences |
| Pseudocode | Yes | Algorithm 1 SPUCB |
| Open Source Code | No | The paper does not provide concrete access to source code. It mentions "Our method has also been deployed as a service to support online businesses in Alibaba." but no public release. |
| Open Datasets | No | The synthetic dataset is randomly generated following our assumptions." and "The real-world dataset is collected from one of the largest ecommercial platform in China for the problem of products recommendation." Neither is publicly available with access info. |
| Dataset Splits | No | The paper mentions "All super-parameters of the above algorithms are tuned by a cross-validation experiment with the best performance" but does not provide specific dataset split information (percentages, sample counts, or citations to predefined splits) for the main experiments. |
| Hardware Specification | No | "The total running time for online updating parameters is less than 20ms with a single machine with 4-core CPUs." This describes the CPU core count but lacks specific hardware model details like CPU model, GPU model, or memory. |
| Software Dependencies | No | The paper does not provide specific ancillary software details with version numbers. |
| Experiment Setup | Yes | All super-parameters of the above algorithms are tuned by a cross-validation experiment with the best performance. (Specifically, UCB: c = 0.2, LINUCB: α = 3.5, LSPS: σ1 = 0.3, σ2 = 0.01, σ3 = 0.3, SPUCB: R = 0.2, δ = 0.9, λ = 1.0, Rr = 0.5) |