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]
DALEX: Explainers for Complex Predictive Models in R
Authors: Przemyslaw Biecek
JMLR 2018 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | This paper describes a consistent collection of explainers for predictive models, a.k.a. black boxes. Each explainer is a technique for exploration of a black box model. Presented approaches are model-agnostic, what means that they extract useful information from any predictive method irrespective of its internal structure. Each explainer is linked with a specific aspect of a model. Some are useful in decomposing predictions, some serve better in understanding performance, while others are useful in understanding importance and conditional responses of a particular variable. Every explainer presented here works for a single model or for a collection of models. In the latter case, models can be compared against each other. Such comparison helps to find strengths and weaknesses of different models and gives additional tools for model validation. Presented explainers are implemented in the DALEX package for R. |
| Researcher Affiliation | Collaboration | Przemys law Biecek EMAIL Faculty of Mathematics and Information Science, Warsaw University of Technology 75 Koszykowa Street, Warsaw, Poland Samsung Research Poland |
| Pseudocode | No | The paper describes the architecture and functionality of the DALEX package and its explainers, but it does not present any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Presented framework is available as an open source package DALEX for R. The R language (R Core Team, 2017) is one of the most popular languages for statistical and machine learning modeling. DALEX works with any predictive model. The extended user documentation1 contains examples for the most popular frameworks, such as caret (Kuhn, 2008), mlr (Bischl et al., 2016), Random Forest and Gradient Boosting Machines. The DALEX package is available on at CRAN and Git Hub2 along with technical documentation3. 2. Development version is available at https://github.com/pbiecek/DALEX. |
| Open Datasets | No | The paper discusses the DALEX package for explaining predictive models and illustrates its functionalities with examples, such as those shown in Figure 1 related to 'Prediction: Price' or 'Variable: Construction.year'. However, it does not explicitly name or provide access information for any specific dataset used in these examples or for evaluating the DALEX package itself. The focus is on the package's capabilities rather than a particular experimental dataset. |
| Dataset Splits | No | The paper introduces a software package for model explanations and showcases its features. It does not describe any specific experiments with datasets that would require detailing training, validation, or test splits. The paper focuses on the functionality of the DALEX explainers rather than specific model training or evaluation protocols. |
| Hardware Specification | No | The paper describes a software package (DALEX for R) and its methodology for explaining predictive models. It does not contain any information regarding the specific hardware (e.g., GPU, CPU models, memory, or cloud resources) used for developing or demonstrating the package's functionalities or for any experiments. |
| Software Dependencies | Yes | Here we presented DALEX 0.2.5 with R 3.5.1. |
| Experiment Setup | No | The paper introduces a framework and an R package (DALEX) for model-agnostic explanations. While it describes various explainers and their applications, it does not detail specific experimental setups, such as hyperparameters, training configurations, or system-level settings for any particular predictive model. The figures illustrate the output of the explainers, not the setup of experiments to generate those outputs. |