Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]
Approximate Inference via Weighted Rademacher Complexity
Authors: Jonathan Kuck, Ashish Sabharwal, Stefano Ermon
AAAI 2018 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experiments demonstrate that we can produce tighter bounds than competing methods in both the weighted and unweighted settings. |
| Researcher Affiliation | Collaboration | Jonathan Kuck Computer Science Department Stanford University EMAIL Ashish Sabharwal Allen Institute for Arti๏ฌcial Intelligence Seattle, WA EMAIL Stefano Ermon Computer Science Department Stanford University EMAIL |
| Pseudocode | Yes | Algorithm 1 Rademacher Estimate of log Z(w) |
| Open Source Code | No | The paper does not provide explicit links or statements regarding the open-source code for the methodology it describes. |
| Open Datasets | Yes | The models used in our experiments can be downloaded from http://reasoning.cs.ucla.edu/c2d/results.html |
| Dataset Splits | No | The paper does not explicitly provide training/test/validation dataset splits, percentages, or sample counts, nor does it refer to predefined splits with citations for reproducibility. |
| Hardware Specification | No | The paper mentions using specific software implementations (python maxflow module, Max HS solver) but does not provide any specific hardware details such as GPU/CPU models or memory used for the experiments. |
| Software Dependencies | No | The paper mentions software like 'python maxflow module' and 'Max HS' but does not specify their version numbers or other software dependencies with versioning for reproducibility. |
| Experiment Setup | Yes | Figure 2: Bounds for a 7x7 spin glass model with k = 5 (for both methods), that hold with probability .95. |