The Apprentice Learner Architecture provides a framework for modeling and simulating learners working educational technologies. There are three general GitHub repositories for the AL Project:
- AL_Core (https://github.com/apprenticelearner/AL_Core), which is the core library for learner modeling used to configure and instantiate agents and author their background knowledge.
- AL_Train (this respository), which contains code for interfacing AL agents with CTAT-HTML tutors and running training experiments.
- AL_Outerloop (https://github.com/apprenticelearner/AL_Outerloop), which provides additional functionality to AL_Train simulating adaptive curricula.
This repository does the following:
- Provides the altrain executable which is used to train Apprentice Learner agents from a training.json spec file, either in batch mode or interactively.
- Implements a host server with a Datashop logging language compatable logger which will write tab-delimited transaction files of AL agent actions.
To install the AL_Train library, first follow the installation instructions for the AL_Core Library. Next, clone the respository to your machine using the GitHub deskptop application or by running the following command in a terminal / command line:
git clone https://github.com/apprenticelearner/AL_TrainNavigate to the directory where you cloned AL_Train in a terminal / command line and run:
python -m pip install -e .Everything should now be fully installed and ready.
- Source code: https://github.com/apprenticelearner/AL_Train
- Documentation: https://al-core.readthedocs.io/en/latest/
We have created a number of examples to demonstrate basic usage of the Appentice Learner that make use of this repository as well as AL_Core. These can be found on the examples page of the AL_Core wiki.
If you use the interactive training components of this software in a scientific publiction, then we would appreciate a citation of the following paper:
Daniel Weitekamp III, Erik Harpstead, and Kenneth R Koedinger. 2020. An Interaction Design for Machine Teaching to Develop AI Tutors. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems - CHI ’20,. https://doi.org/10.1145/3313831.3376226
Bibtex entry:
@inproceedings{WeitekampIII2020,
author = {{Weitekamp III}, Daniel and Harpstead, Erik and Koedinger, Kenneth R},
booktitle = {Proceedings of the SIGCHI Conference on Human Factors in Computing Systems - CHI '20,},
doi = {10.1145/3313831.3376226},
file = {:C$\backslash$:/Users/eharpste/Documents/Articles/Weitekamp III, Harpstead, Koedinger - 2020 - An Interaction Design for Machine Teaching to Develop AI Tutors.pdf:pdf;:C$\backslash$:/Users/eharpste/Documents/Articles/Weitekamp III, Harpstead, Koedinger - 2020 - An Interaction Design for Machine Teaching to Develop AI Tutors(2).pdf:pdf},
isbn = {9781450367080},
keywords = {"Simulated Learners,Intelligent Tutoring Systems",Interaction Design,Machine Teaching,Programming-by-Demonstration},
title = {{An Interaction Design for Machine Teaching to Develop AI Tutors}},
year = {2020}
}
If you use the broader Apprentice Learner Architecture in a scientific publication, then we would appreciate a citation of the following paper:
Christopher J MacLellan, Erik Harpstead, Rony Patel, and Kenneth R Koedinger. 2016. The Apprentice Learner Architecture: Closing the loop between learning theory and educational data. In Proceedings of the 9th International Conference on Educational Data Mining - EDM ’16, 151–158. Retrieved from http://www.educationaldatamining.org/EDM2016/proceedings/paper_118.pdf
Bibtex entry:
@inproceedings{MacLellan2016a,
author = {MacLellan, Christopher J and Harpstead, Erik and Patel, Rony and Koedinger, Kenneth R},
booktitle = {Proceedings of the 9th International Conference on Educational Data Mining - EDM '16},
pages = {151--158},
title = {{The Apprentice Learner Architecture: Closing the loop between learning theory and educational data}},
url = {http://www.educationaldatamining.org/EDM2016/proceedings/paper{\_}118.pdf},
year = {2016}
}