Automated Design of Robust Mechanisms
Authors: Michael Albert, Vincent Conitzer, Peter Stone
AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In Figure 1, we show the performance of the optimal ex-post, robust, and Bayesian mechanisms using our estimated distribution as we increase the number of samples. We report confidence intervals for both the ex-post mechanisms and the robust mechanisms... |
| Researcher Affiliation | Academia | Michael Albert, Vincent Conitzer Department of Computer Science Duke University Durham, NC 27708, USA {malbert,conitzer}@cs.duke.edu Peter Stone Department of Computer Science University of Texas at Austin Austin, TX 78712, USA pstone@cs.utexas.edu |
| Pseudocode | No | The paper describes the computational methods and linear programs in prose but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement regarding the release of open-source code for the methodology described, nor does it include a link to a code repository. |
| Open Datasets | No | The paper states, 'We model the true distribution as a categorical distribution with 10 x 10 elements, with each element corresponding to a tuple (θ, ω)... we use a discretized bi-variate normal distribution.' and 'To estimate the distribution, we sample from the true distribution'. This indicates they generated their own data rather than using a publicly available dataset, and no access information is provided. |
| Dataset Splits | No | The paper describes generating data by sampling from a discretized bi-variate normal distribution and estimating it using Bayesian updating. It discusses the 'number of samples' used for estimation, but does not define traditional training, validation, and test dataset splits with percentages or counts. |
| Hardware Specification | No | The paper does not specify any hardware details such as GPU/CPU models, memory, or specific computing environments used for running the experiments. |
| Software Dependencies | No | The paper does not provide any specific software dependencies or version numbers (e.g., programming languages, libraries, or solvers) used for the implementation or experiments. |
| Experiment Setup | Yes | Throughout the experiments, we have a single bidder with type θ {1, 2, ..., 10} and valuation v(θ) = θ. The external signal is ω {1, 2, ..., 10}. We model the true distribution as a categorical distribution with 10 × 10 elements... We use a discretized bi-variate normal distribution... To estimate the distribution, we sample from the true distribution and use Bayesian updating with a maximally uninformative Dirichlet prior (α = [1, ..., 1])... The parameters used were as follows: Correlation = .5, ϵ = .05. Each experiment was repeated 200 times... |