Blackbox AI: Developing Trust and Explain-ability for Responsible AI
15 Jun 2023
In both public and private sectors, companies won't implement AI over a lack of trust and fear exposure to risk. Are your practitioners struggling to understand how their model actively builds to results and therefore how they think and process? In this session, we'll explore the fundamentals of explain-ability through the eyes of post-hoc explainability and other methods that fight the black box nature of many neural networks. We'll also touch upon transparency and how this plays into achieving the same objectives.