This Project was Developed as part of the Doctoral Dissertation in the University of Urbino, Italy.

Toward Enhanced Cognitive Learning with Explainable AI (CL-XAI)

Welcome to the GitHub repository for CL-XAI, an interactive game designed to enhance cognitive learning through domain-agnostic combinatorial tasks. The game is deployed on our web server, and you can access it here. The open source code is availabel at the following GiHub repository ExpGame.

Overview

As artificial intelligence (AI) systems become increasingly prevalent in making critical decisions, there is a growing demand for Explainable AI (XAI). CL-XAI addresses this demand by providing insights into the internal mechanisms of AI models, fostering cognitive learning and understanding of complex concepts.

Key Objectives

Our research focuses on two main objectives:

1. Understanding Human Comprehension: Explore how human learners comprehend AI models using XAI tools.
2. Evaluating Tool Effectiveness: Evaluate the effectiveness of XAI tools through valuable human feedback.

CL-XAI in Action

CL-XAI is demonstrated through a game-inspired virtual use case. In this scenario, learners engage in combinatorial problem-solving tasks. The goal is to enhance problem-solving skills and deepen understanding of complex concepts. This innovative approach showcases the transformative potential of CL-XAI in the field of cognitive learning and co-learning.

Getting Started

To experience CL-XAI, simply click here to access the interactive game.

Further Reading

We appreciate your interest in CL-XAI and invite you to explore, contribute, and provide feedback. Together, let's advance the frontiers of cognitive learning with Explainable AI!

For further reading about this project and related publication please visit: Toward enriched Cognitive Learning with XAI.

Our other related works:

• FCE: Feedback Based Counterfactual Explanations for Explainable AI

• Investigating Human-Centered Perspectives in Explainable Artificial Intelligence

• Towards Human Cognition Level-based Experiment Design for Counterfactual Explanations

• How to Build Self-Explaining Fuzzy Systems: From Interpretability to Explainability [AI-eXplained]

Announcement

The related code for reproducing the experiments will be disclosed soon as the paper is in publishing phase.

Cite As

If you use CL-XAI in your research, please cite it as:

@article{Suffian2023,
  title={Toward Enriched Cognitive Learning with XAI},
  author={Muhammad Suffian and Ulrike Kuhl and Jose M. Alonso-Moral and Alessandro Bogliolo},
  year={2023},
  eprint={2312.12290},
  archivePrefix={arXiv},
  primaryClass={cs.AI}
}

License

This work is licensed under a Creative Commons Attribution 4.0 International License.