Online Budgeted Matching with General Bids
Authors: Jianyi Yang, Pengfei Li, Adam Wierman, Shaolei Ren
NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We validate the empirical benefits of Meta Ad and LOBM through numerical experiments on the applications of an online movie matching an VM placement on physical servers. |
| Researcher Affiliation | Academia | Jianyi Yang University of Houston Houston, TX, USA jyang71@central.uh.edu Pengfei Li University of California, Riverside Riverside, CA, USA pli081@ucr.edu Adam Wierman California Institute of Technology Pasadena, CA, USA adamw@caltech.edu Shaolei Ren University of California, Riverside Riverside, CA, USA shaolei@ucr.edu |
| Pseudocode | Yes | Algorithm 1 Meta Algorithm (Meta Ad) ... Algorithm 5 Learning-Augmented OBM (LOBM, w/o FLM) |
| Open Source Code | No | We will open source the codes upon the publication of the paper. |
| Open Datasets | Yes | We run the online movie matching application based on a real dataset of Movie Lens [12]. |
| Dataset Splits | Yes | We generate 10k, 1k, and 1k samples of graph instances based on the Movie Lens dataset for training, validation and testing, respectively. |
| Hardware Specification | No | The training process on a laptop takes around 1 hour, while the inference process over each instance takes less than one second. ... The experiments can be reproduced by a personal computer with CPU. |
| Software Dependencies | Yes | The neural networks are trained by Adam optimizer with a learning rate of 10 3 for 50 epochs. The offline optimal solution is obtained using Gurobi [11]. |
| Experiment Setup | Yes | The optimal parameter θ governing the level of conservativeness in Meta Ad is tuned based on the validation dataset. ... The neural network has two layers, each with 200 hidden neurons. The neural networks are trained by Adam optimizer with a learning rate of 10 3 for 50 epochs. |