4 ms·
>Our key insight is that the diagonal linear recurrent layer can act as a gradient accumulator So they're sort of reinventing the discrete-time differentiator
by eli_gottlieb 2y ago
>Our key insight is that the diagonal linear recurrent layer can act as a gradient accumulator
So they're sort of reinventing the discrete-time differentiator from signal processing, but parameterized neurally?
- radarsat1 2y agoConverging slowly on Kalman filters, calling it now.