Using tools that increase the explainability of ML systems is one way to develop responsible and explainable AI. However, with such a large number of ethical AI tools available, many practitioners are at a loss for how to decide which is the best fit for their product/system. To better understand the landscape, PAI’s ABOUT ML team conducted a qualitative research study focused on the scope and features of existing XAI tools. Through this work, PAI has developed an XAI Toolsheet to compare tools easily.
The ABOUT ML team at Partnership on AI is hosting a virtual workshop on July 12th exploring XAI tools and how to evaluate their features based on our research.
Through this workshop, participants will:
- Receive a first look at PAI’s XAI toolsheet developed by PAI Research Fellow Surya Karunagaran that provides an initial set of evaluation factors for practitioners to consider when choosing the XAI tool for their project
- Engage in activities designed to generate feedback on PAI’s evaluation framework for XAI tools before publication.
This workshop is intended for:
- Tools users: Individuals such as AI/ML practitioners, data scientists, engineers, product developers, and product managers who are seeking to use XAI tools in their work
- Tool developers: Individuals who design, develop, and maintain XAI tools
- Academic researchers: Individuals who work on topics related to XAI tools and documentation
If you’re interested in joining, please email sarah.v@partnershiponai.org