DAGs with NO TEARS: Continuous Optimization for Structure Learning
Authors: Xun Zheng, Bryon Aragam, Pradeep K. Ravikumar, Eric P. Xing
NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We compared our method against greedy equivalent search (GES) [9, 31], the PC algorithm [42], and Li NGAM [38]. ... In each experiment, a random graph G was generated from one of two random graph models, Erdös Rényi (ER) or scale-free (SF). ... We now examine our method for structure recovery, which is shown in Figure 3. ... In our experiments we generated random graphs with d = 10, and then generated 10 simulated datasets containing n = 20 samples (for high-dimensions) and n = 1000 (for low-dimensions). We then compared the scores returned by our method to the exact global minimizer computed by GOBNILP along with the estimated parameters. The results are shown in Table 1. |
| Researcher Affiliation | Collaboration | Xun Zheng1, Bryon Aragam1, Pradeep Ravikumar1, Eric P. Xing1,2 1Carnegie Mellon University 2Petuum Inc. {xunzheng,naragam,pradeepr,epxing}@cs.cmu.edu |
| Pseudocode | Yes | Algorithm 1 NOTEARS algorithm |
| Open Source Code | Yes | The implementation is publicly available at https://github.com/xunzheng/ notears. |
| Open Datasets | Yes | In each experiment, a random graph G was generated from one of two random graph models, Erdös Rényi (ER) or scale-free (SF). ... We also compared FGS and NOTEARS on a real dataset provided by Sachs et al. [33]. |
| Dataset Splits | No | The paper does not explicitly state the use of validation splits or a specific validation methodology like cross-validation. It mentions training models and evaluating on test sets but lacks details on validation splits. |
| Hardware Specification | No | No specific hardware details (e.g., GPU/CPU models, memory, or cloud instances) used for running experiments were mentioned in the paper. |
| Software Dependencies | No | The paper mentions L-BFGS, PQN method, and GOBNILP but does not specify version numbers for these or any other software components. |
| Experiment Setup | Yes | For brevity, we outline the basic set-up of our experiments here; precise details of our experimental set-up, including all parameter choices and more detailed evaluations, can be found in Appendix E. |