BayesiaLab 9.0: New Features & Updates (12/2019)
Here is a small selection of new or updated features released in BayesiaLab 9.0:
- The Target and Function nodes optimization tools have been enhanced with new options and outputs.
- Get precise and concise explanations for your current set of evidence with the Most Relevant Explanations function.
- Incorporate your partial prior knowledge on the structure with Structural Priors.
- Try to improve the quality of the machine learned models with the new BayesiaLab's Smoothed Bootstrapping algorithm, Data Perturbation, that perturbs your data not only with the weight of each particle but also with the overall Structural Coefficient.
- Automatically estimate Structural Priors via Resampling/Bagging. This is particularly powerful for small data sets since you no longer need to search for the best Structural Coefficient.
- Induce a Partial Order among your variables via Resampling/Bagging, a step toward learning causal Bayesian networks.
- The Markov Blanket Learning Algorithms can now take into account constraints on arc directionality expressed with Temporal Indices or Forbidden Arcs.
- The cross-validation of Variable Clustering features now Purities to estimate to quality of the Factors.
- The code to compute the posterior probability of the Target node given its Markov Blanket can now be Exported in Python (available by subscription only).
- The presentation slides and recorded videos of the 3-Day Introductory and Advanced Courses are now available via the new Media tool (available by subscription only).
- Use the new textures of the 3D-Mapping tool and its auto-rotate tool for creating nice presentations of your models.