Exact MAP Inference by Avoiding Fractional Vertices
Authors: Erik M. Lindgren, Alexandros G. Dimakis, Adam Klivans
ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We consider a synthetic experiment on randomly created graphical models, which were also used in (Sontag & Jaakkola, 2007; Weller, 2016; Weller et al., 2014). The graph topology used is the complete graph on 12 nodes. We first reparametrize the model to use the sufficient statistics... and We experimentally verify this condition and demonstrate how efficient various integer programming methods are at removing fractional solutions. |
| Researcher Affiliation | Academia | 1Department of Electrical and Computer Engineering, University of Texas at Austin, USA 2Department of Computer Science, University of Texas at Austin, USA. |
| Pseudocode | Yes | Algorithm 1 Branch and Bound and Algorithm 2 M-best Integral and Algorithm 3 Estimate S(VC) for Binary, Pairwise Graphical Models |
| Open Source Code | No | The paper does not provide concrete access to source code (specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described in this paper. |
| Open Datasets | No | The paper mentions 'randomly created graphical models' and cites previous work where similar models were used, but does not provide concrete access information (link, DOI, repository, or formal citation with authors/year for a specific publicly available dataset instance) for the data used in their experiments, nor does it state that their generated data is publicly available. |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers like Python 3.8, CPLEX 12.4) needed to replicate the experiment. |
| Experiment Setup | Yes | The node weights are drawn θi Uniform( 1, 1) and the edge weights are drawn Wij Uniform( w, w) for varying w. The quantity w determines how strong the connections are between nodes. We do 100 draws for each choice of edge strength w. |