Study: Deep neural networks don't see the world the way we d

Computational models of hearing and vision can build up their own idiosyncratic “invariances” — meaning they respond the same way to stimuli with very different features, according to MIT scientists.

Related Keywords

Guillaume Leclerc , Josh Mcdermott , Flatiron Institute Center , Department Of Energy Computational Science Graduate Fellowship , Friends Of The Mcgovern Institute Fellowship , National Science Foundation , National Institutes Of Health , Mcgovern Institute For Brain Research , Govern Institute , Brain Research , Feather Phd , Computational Neuroscience , Cadence Design Systems Professor , National Institutes , Energy Computational Science Graduate Fellowship , Mcgovern Institute , Deep Neural Networks , Sensory Perception , Vision , Hearing , Invariance , Retamers , Enelle Feather ,