Results from the Probablistic Inference Evaluation of UAI'08
A. Darwiche, R. Dechter, A. Choi, V. Gogate, and L. Otten
This page describes the Probabilistic Inference Evaluation that was held at UAI 2008. The original motivation, goals, and setup are described in the workshop CFP and the evaluation CFP. Below we give the details of the evaluation setup and the results obtained.
The organizers wish to thank all the teams and their members for participating in the evaluation. All teams are to be commended as the success of this evaluation was crucially dependent on each and every one of the teams. All the teams did very well (as the results below show). The co-chairs also wish to thank the significant help provided by the graduate student organizers.
The presentation given at UAI'08 is also available here as ppt slides. The information in the presentation is also organized and elaborated on in the pages below.
The Evaluation Setup
Evaluation description: This is the original call for participation, which includes the setup of the evaluation and the task definitions.
Evaluation environment: This describes the system on which the evaluation was performed.
Benchmarks and benchmark descriptions: This page contains the Bayesian and Markov networks used in the evaluation, available for download, with summary descriptions.
Participation
Solver descriptions: a list of participants and solvers evaluated.
Results
MPE results
Exact PR (probability of evidence) results
Exact MAR (variable marginals) results
Approximate PR (probability of evidence) results
Approximate MAR (variable marginals) results
Organization
Previous Evaluations
The first evaluation of probabilistic reasoning systems was presented at UAI'06.