EthicalML/xai

By Stephen Downes - Stephen's Web. EthicalML/xai
GitHub, 2019/03/22
From the website: " XAI is a Machine Learning library that is designed with AI explainability in its core. XAI contains various tools that enable for analysis and evaluation of data and models. The XAI library is maintained by The Institute for Ethical AI & ML, and it was developed based on the 8 principles for Responsible Machine Learning." This may all seem pretty esoteric, but I can imagine a  student assignment of the (far) future: use your AI to automatically generate a response to a Reddit discussion and them employ the AI to explain the reasoning for your AI-generated response. More...