Roadmap

Core

The list of items shown below are likely to be implemented due to their mainstream use, ability to increase learning speed, or ability to reduce computational resources.

• Online Decision Trees And Boosting Algorithms

  • Currently, the offline variants offer superior performance in terms of generalization for tabular datasets. However, because they tend to be computationally expensive and requires a full dataset, it is not suitable for real-time game environments.

  • The research literature on the online variants of these algorithms are lacking, and so we are waiting for more papers to come out.

  • Additionally, we lack experience in developing decision trees and boosting algorithms, which may result in long development times of these algorithms.

• Incremental DBSCAN

  • It is an online version of DBSCAN that allows it to construct clusters from individual datapoints.

• Online EM

  • It is an online version of EM that allows it to construct clusters from individual datapoints.

Nice-To-Have

The list of items shown below may not necessarily be implemented in the future. However, they could be prioritized with external demand, collaboration, or funding.

• None