Structured Features in Naive Bayes Classification
Authors: Arthur Choi, Nazgol Tavabi, Adnan Darwiche
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluated our SNB classifier based on independent testing sets, using simulated games for the players of each pair of playing styles. First, consider the SNB classifier with PSDDs (solid lines). As we increase the number of games observed, we get more accurate classifications of a player s skill. Here, we see that a beginner, who plays randomly, is easiest to classify, requiring only a few games to obtain a high level of accuracy. |
| Researcher Affiliation | Academia | Arthur Choi Computer Science Department University of California, Los Angeles aychoi@cs.ucla.edu Nazgol Tavabi Department of Computer Engineering Sharif University of Technology ntavabi@ce.sharif.edu Adnan Darwiche Computer Science Department University of California, Los Angeles darwiche@cs.ucla.edu |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper mentions external tools and libraries like 'publicly available SDD compiler at http://reasoning.cs.ucla.edu/sdd' and the 'GRAPHILLION library', but it does not provide open-source code for the SNB classifier itself or its implementation. |
| Open Datasets | No | The paper states, 'To obtain tic-toe-toe games of varying but known styles, we simulated games from different tic-tac-toe programs.' and 'We simulated training sets... for the normal distribution.' This indicates data was simulated/generated, not sourced from a publicly available, citable dataset with access information. |
| Dataset Splits | No | The paper mentions 'training sets' and 'testing datasets' but does not specify validation splits, exact percentages for splits, or reference to predefined standard splits with citations. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions using a 'publicly available SDD compiler' and the 'GRAPHILLION library' but does not provide specific version numbers for these or any other software dependencies. |
| Experiment Setup | Yes | For learning PSDDs, we assumed Dirichlet priors with exponents 2 (corresponding to Laplace smoothing). |