Scalable Robust Kidney Exchange
Authors: Duncan C McElfresh, Hoda Bidkhori, John P Dickerson1077-1084
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we compare each robust formulation against the leading non-robust formulation, PICEF (Dickerson et al. 2016), with varying levels of uncertainty. These experiments use real exchange graphs collected from the United Network for Organ Sharing (UNOS) a large US-wide kidney exchange with over 160 participating transplant centers between 2010 and 2016, as well simulated exchanges generated from known patient statistics using the standard method (Dickerson, Procaccia, and Sandholm 2018). |
| Researcher Affiliation | Academia | Duncan C Mc Elfresh Department of Mathematics University of Maryland College Park, MD 20742 dmcelfre@math.umd.edu Hoda Bidkhori Department of Industrial Engineering University of Pittsburgh Pittsburgh, PA 15260 bidkhori@pitt.edu John P Dickerson Department of Computer Science University of Maryland College Park, MD 20742 john@cs.umd.edu |
| Pseudocode | No | The paper presents mathematical formulations (e.g., Problem (1), (2), (4a)-(4r)), but does not contain any pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | Yes | Our code is available on Git Hub: https://github.com/duncanmcelfresh/RobustKidneyExchange. |
| Open Datasets | Yes | These experiments use real exchange graphs collected from the United Network for Organ Sharing (UNOS) a large US-wide kidney exchange with over 160 participating transplant centers between 2010 and 2016, as well simulated exchanges generated from known patient statistics using the standard method (Dickerson, Procaccia, and Sandholm 2018). |
| Dataset Splits | No | The paper does not provide explicit training, validation, or test dataset splits. It mentions using 'real exchange graphs' and 'simulated exchanges' and drawing 'realizations', but does not specify how these datasets are partitioned into splits for training, validation, or testing. |
| Hardware Specification | No | The paper states, 'All experiments were implemented in Python and used Gurobi (Gurobi Optimization, Inc. 2018), a state-of-the-art industrial combinatorial optimization toolkit, as a sub-solver,' but does not provide specific details about the hardware (e.g., CPU, GPU models, RAM) used for running the experiments. |
| Software Dependencies | Yes | All experiments were implemented in Python and used Gurobi (Gurobi Optimization, Inc. 2018), a state-of-the-art industrial combinatorial optimization toolkit, as a sub-solver. |
| Experiment Setup | Yes | We compute MR for protection levels ϵ {10 4, 10 3, 10 2, 10 1, 0.5}, and then calculate both OPT (MR) and OPT (MNR). Figure 2 shows OPT on realistic 64-vertex simulated graphs (left) and larger (typically 150 300-vertex) real UNOS graphs (right); these figures show results for each protection level ϵ and for various α. ... Each Γ corresponds to a different notion of uncertainty, such that exactly Γ edges fail. |