I’d be curious whether Chris and the other members of the Clarity team have thought about (in their most optimistic vision) how “microscope AI” could enable other sciences—biology, neuroscience, physics, etc. -- and if so whether they’d be willing to share. AIUI, the lack of interpretability of neural networks is much more of an immediate issue for trying to use deep learning in these areas because just being able to predict tends to be less useful.
I’d be curious whether Chris and the other members of the Clarity team have thought about (in their most optimistic vision) how “microscope AI” could enable other sciences—biology, neuroscience, physics, etc. -- and if so whether they’d be willing to share. AIUI, the lack of interpretability of neural networks is much more of an immediate issue for trying to use deep learning in these areas because just being able to predict tends to be less useful.