1 paper · 1 filter
Hans Dermot Doran, Suzana Veljanovska
Making AI safe and dependable requires the generation of dependable models and dependable execution of those models. We propose redundant execution as a well-known technique that c…