When AI can explain to us that it's unsure, it adds a critical layer of transparency for its safe deployment and use. We’re developing ways to foster and streamline the common practices of quantifying, evaluating, improving, and communicating uncertainty in the AI application development lifecycle.
Tools + code
Dennis Wei2021ICML 2021
Joshua Lee, Yuheng Bu, et al.2021ICML 2021
Umang Bhatt, Javier Antorán, et al.2021AIES 2021
Benjamin Elder, Matthew Arnold, et al.2021AAAI 2021
Soumya Ghosh, William T. Stephenson, et al.2020NeurIPS 2020
Raphaël Pestourie, Youssef Mroueh, et al.2020npj Computational Materials