Selecting Compliant Agents for Opt-in Micro-Tolling
Authors: Josiah P. Hanna, Guni Sharon, Stephen D. Boyles, Peter Stone565-572
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experimental results on three real-world traffic scenarios suggest that evaluating the marginal impact of a given agent serves as a particularly strong heuristic for selecting an agent to be compliant. Results from using this heuristic for selecting 7.6% of the agents to be compliant achieved an increase of up to 10.9% in social welfare over not tolling at all. We present experimental results obtained from a dynamic traffic assignment simulation of three real-world traffic scenarios. |
| Researcher Affiliation | Academia | Josiah P. Hanna,1 Guni Sharon,2 Stephen D. Boyles,1 Peter Stone1 1 The University of Texas at Austin 2 Texas A & M University |
| Pseudocode | No | No structured pseudocode or algorithm blocks were found in the paper. |
| Open Source Code | No | The paper does not provide any concrete access information (link, explicit statement of release) to the source code for the methodology described. |
| Open Datasets | Yes | The traffic scenarios are available online at: https://goo.gl/SyvV5m. |
| Dataset Splits | No | The paper describes using three traffic scenarios (Sioux Falls, Austin, San Antonio) with specific network and demand table sizes. However, it does not provide explicit details about traditional training, validation, and test dataset splits in terms of percentages or sample counts, as it is a simulation study rather than a typical machine learning model training setup. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used to run the experiments (e.g., CPU/GPU models, memory specifications). |
| Software Dependencies | No | The paper mentions using a 'Dynamic Traffic Assignment (DTA) simulator (Levin and Boyles 2015b)' and modeling traffic through the 'cell transmission model (CTM)'. However, it does not specify any software names with version numbers for dependencies or environments. |
| Experiment Setup | Yes | The top row of Figure 2 shows results for each heuristic as we vary the compliance level with R = 1 10 4, β = 4 (the parameter settings used by Sharon et al. (2017b)). We set β = 4 in all experiments. For each compliance level and heuristic method we run 20 trials and average the resulting total social welfare values. The DMCP+TE heuristic function is defined as: h DMCP+TE = (1 α)h DMCP + αh TE(a) where α is a small, positive constant (we use 0.01). If a vehicle is unable to enter a link because the link s receiving flow is zero for more than 96 seconds (16 time steps), the vehicle attempts to reroute itself through the least cost path leading to its destination that avoids the jammed link. |