During my undergraduate studies at the University of Waikato, I built a system for performing Linear Genetic Programming (LGP). The system was developed to offer an open-source implementation of LGP and was submitted as my Honours project.
LGP is a paradigm of genetic programming that employs a representation of linearly sequenced instructions in automatically generated programs. A population of imperative style programs are trained on a particular dataset and the best resulting program can be used as a predictor for the problem at hand.
The system performs well overall, but one issue that had always bugged me was that training was performed synchronously and would block until completion,... Continue Reading...
|Date||October 14, 2018|
|Filename||2018-10-14-LGP - Asynchronous Training.md|
|Last Modified||Thu, 04 Jun 2020|
|Summary||Details about the implementation of asynchronous training in the LGP system using Kotlin coroutines|
|Tags||lgp, kotlin, coroutines, asynchronous, training, async|
|Title||LGP - Asynchronous Training using Kotlin Coroutines|